# Physics-guided deep learning

A deep-learning approach that integrates first-principled physical knowledge into data-driven methods.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## {\\lambda}Split: Self-Supervised Content-Aware Spectral Unmixing for Fluorescence Microscopy

DevFeed: [{\\lambda}Split: Self-Supervised Content-Aware Spectral Unmixing for Fluorescence Microscopy](<https://devfeed.tech/articles/lambda-split-self-supervised-content-aware-spectral-unmixing-for-fluorescence-microscopy-144881.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2603.23647>)

Author: Federico Carrara, Talley Lambert, Mehdi Seifi, Florian Jug

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Physics-guided deep learning](<https://devfeed.tech/topics/physics-guided-deep-learning.md>), [Image](<https://devfeed.tech/topics/image.md>)

Tags: [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [image](<https://devfeed.tech/tags/image.md>), [learning](<https://devfeed.tech/tags/learning.md>), [microscopy](<https://devfeed.tech/tags/microscopy.md>), [model](<https://devfeed.tech/tags/model.md>), [noise](<https://devfeed.tech/tags/noise.md>), [overlap](<https://devfeed.tech/tags/overlap.md>)

### AI overview

λSplit addresses fluorescence spectral unmixing, where pixel-wise least-squares methods can degrade with overlapping emissions and noise. It uses a hierarchical Variational Autoencoder to learn a conditional distribution over fluorophore concentration maps, with a differentiable Spectral Mixer enforcing consistency with image formation. The authors report evaluation on 66 challenging benchmarks derived from three real-world datasets, compared with 10 classical and learning-based baselines. They report competitive performance and greater robustness under high noise, substantial spectral overlap, and reduced spectral dimensionality, as well as implicit noise removal. The method is described as compatible with standard confocal microscope data without hardware modifications.

### Source excerpt

arXiv:2603.23647v4 Announce Type: replace-cross Abstract: In fluorescence microscopy, spectral unmixing aims to recover individual fluorophore concentrations from spectral images that capture mixed fluorophore emissions. Since classical methods operate pixel-wise and rely on least-squares fitting, their performance degrades with increasingly overlapping emission spectra and higher levels of noise, suggesting that a data-driven approach that can learn and utilize a structural prior might lead to improved results. Learning-based approaches for spectral imaging do exist, but they are either not optimized for microscopy data or are developed for very specific cases that are not applicable to fluorescence microscopy settings. To address this, we propose {\lambda}Split, a physics-informed deep generative model that learns a conditional distribution over concentration maps using a hierarchical Variational Autoencoder. A fully differentiable Spectral Mixer enforces consistency with the image formation process, while the learned structural priors enable state-of-the-art unmixing and implicit noise removal. We demonstrate {\lambda}Split on 3 real-world datasets that we synthetically cast into a total of 66 challenging spectral unmixing benchmarks. We compare our results against a total of 10 baseline methods, including classical methods and a range of learning-based methods. Our results consistently show competitive performance and improved robustness in high noise regimes, when spectra overlap considerably, or when the spectral dimensionality is lowered, making {\lambda}Split a new state-of-the-art for spectral unmixing of fluorescent microscopy data. Importantly, {\lambda}Split is compatible with spectral data produced by standard confocal microscopes, enabling immediate adoption without specialized hardware modifications.

## Uncertainty-aware damage identification in short-span bridges via physics-informed variational autoencoder

DevFeed: [Uncertainty-aware damage identification in short-span bridges via physics-informed variational autoencoder](<https://devfeed.tech/articles/uncertainty-aware-damage-identification-in-short-span-bridges-via-physics-informed-variational-autoencoder-143287.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2607.05025>)

Author: Ana Fernandez-Navamuel, A. Javier Omella, Diego Zamora-Sanchez, David Pardo

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [VAE](<https://devfeed.tech/topics/vae.md>), [Physics-guided deep learning](<https://devfeed.tech/topics/physics-guided-deep-learning.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Finite Element Method (FEM)](<https://devfeed.tech/topics/finite-element-method.md>)

Tags: [autoencoder](<https://devfeed.tech/tags/autoencoder.md>), [bridge](<https://devfeed.tech/tags/bridge.md>), [bridges](<https://devfeed.tech/tags/bridges.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [health-monitoring](<https://devfeed.tech/tags/health-monitoring.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [physics-informed](<https://devfeed.tech/tags/physics-informed.md>), [probabilistic](<https://devfeed.tech/tags/probabilistic.md>), [uncertainty](<https://devfeed.tech/tags/uncertainty.md>), [uncertainty-quantification](<https://devfeed.tech/tags/uncertainty-quantification.md>)

### AI overview

The proposed physics-informed Gaussian copula variational autoencoder estimates bridge damage and diagnostic uncertainty from vibration data. It embeds a numerical eigenvalue solver as its decoder, so posterior samples satisfy structural dynamics equations, and uses a Gaussian copula to capture correlations among stiffness reduction factors with fewer parameters than Gaussian mixtures. In a numerical short-span bridge study with elastic bearings, modeling errors, measurement noise, and a finer test discretization, its damage posteriors closely matched sequential Monte Carlo estimates. The method can indicate damage location while representing uncertainty, including cases where noise masks light damage or modal properties are weakly sensitive to damage.

### Source excerpt

arXiv:2607.05025v3 Announce Type: replace Abstract: Vibration-based damage identification in civil infrastructure is a challenging, ill-posed inverse problem due to measurement noise, sparse sensor arrays, and modeling errors. While deep learning is powerful for system identification, deterministic approaches lack reliable uncertainty quantification and can yield physically inconsistent results. This work proposes a probabilistic Scientific Machine Learning framework for structural health monitoring: a physics-informed Gaussian copula variational autoencoder. The methodology introduces two main contributions. First, a numerical eigenvalue solver is embedded directly into the variational autoencoder as its decoder, which replaces data-driven surrogates, ensures that every posterior sample satisfies the governing equations of structural dynamics, and reduces the number of trainable parameters. Second, a Gaussian copula describes the posterior distribution of the stiffness factors, capturing cross-correlations between components with fewer parameters than Gaussian mixtures in real time. We validate the method numerically on a short-span bridge supported on elastic bearings, whose response is dominated by vertical bending, accounting for modeling errors, measurement noise, and a finer finite element discretization for testing. The method estimates the posterior distribution of the damage condition, i.e., the stiffness reduction factors of the deck zones and bearings, and these posteriors closely match the exact ones computed with sequential Monte Carlo sampling. The results reveal that the proposed methodology offers a robust and interpretable decision-making tool, able not only to indicate the presence and location of damage, but also to inform on the uncertainty associated with the diagnosis, for instance, when light damage is masked by the effect of noise or when the measured modal properties present limited sensitivity to damage.

## Confidence-Based Materials Data Quality Assessment via Domain Knowledge and Graph-Structured Learning

DevFeed: [Confidence-Based Materials Data Quality Assessment via Domain Knowledge and Graph-Structured Learning](<https://devfeed.tech/articles/confidence-based-materials-data-quality-assessment-via-domain-knowledge-and-graph-structured-learning-142639.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.23342>)

Author: Sihan Liaing, Ji Zhang, Kehan Wang, Zihao Gao, Wei Ma, Yangang Wang, Zongguo Wang

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Data Quality](<https://devfeed.tech/topics/data-quality.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [outlier-detection](<https://devfeed.tech/topics/outlier-detection.md>), [Physics-guided deep learning](<https://devfeed.tech/topics/physics-guided-deep-learning.md>)

Tags: [cond-mat-mtrl-sci](<https://devfeed.tech/tags/cond-mat-mtrl-sci.md>), [confidence](<https://devfeed.tech/tags/confidence.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [data-quality](<https://devfeed.tech/tags/data-quality.md>), [graph](<https://devfeed.tech/tags/graph.md>), [learning](<https://devfeed.tech/tags/learning.md>), [materials](<https://devfeed.tech/tags/materials.md>), [quality-assessment](<https://devfeed.tech/tags/quality-assessment.md>)

### AI overview

The proposed framework scores materials data quality as continuous sample confidence, combining a nine-dimensional feature-quality prior calibrated with domain rules and statistical evidence, duplicate-aware initialization, sample-feature graph propagation, weakly supervised graph attention, and Memory-GAT historical reference scoring. On 23 ODDS datasets, its generic module reports mean AUC-ROC of 0.866. In repeated mp_20 experiments, rule calibration raised HistGradientBoostingRegressor test R² from 0.538±0.008 to 0.616±0.014; graph refinement raised average precision from 0.943±0.004 to 0.953±0.004. Rule-based screening raised R² from 0.590±0.053 to 0.880±0.027 in MACE subset experiments. These are reported experimental results; the supplied text does not establish broader performance.

### Source excerpt

arXiv:2611.23342v1 Announce Type: cross Abstract: The rapid development of materials machine learning relies on large-scale, heterogeneous datasets, yet model reliability is often compromised by imperfect data from diverse experiments, simulations, and database integration. Real-world materials databases may contain measurement errors, failed calculations, missing or duplicate records, and hidden violations of physicochemical constraints. Such problems are difficult to identify because corrupted samples may remain statistically plausible rather than appearing as explicit outliers. In this paper, we propose a confidence-based materials data quality assessment framework that integrates domain knowledge with graph-structured learning. Data quality is formulated as a continuous confidence estimation problem rather than a binary anomaly identification task, enabling sample-level assessment and historical-reference-based batch comparison. A nine-dimensional feature-quality prior and duplicate-aware sample initialization are calibrated using materials domain rules and statistical evidence. Confidence is further refined through sample-feature bipartite graph propagation and weakly supervised graph attention, while a Memory-GAT module provides historical reference scoring. Experiments on 23 ODDS benchmark datasets show that the generic module achieves an average AUC-ROC of 0.866. In five repeated mixed-contamination experiments on mp\_20, rule calibration improves HistGradientBoostingRegressor test $R^2$ from $0.538\pm0.008$ to $0.616\pm0.014$. Graph refinement increases average precision from $0.943\pm0.004$ to 0.953\pm0.004$, while the rule-calibrated configuration has the highest mean test $R^2$ among the evaluated screening configurations. Five repeated MACE subset experiments further show an increase in $R^2$ from $0.590\pm0.053$ to $0.880\pm0.027$ after rule-based screening.

## Uncovering the Invisible in Routine Radiology: Cross-Modal Detection of MRI-Verified Soft-Tissue Pathologies from CBCT via Physics-Guided 3D Self-Supervised Learning

DevFeed: [Uncovering the Invisible in Routine Radiology: Cross-Modal Detection of MRI-Verified Soft-Tissue Pathologies from CBCT via Physics-Guided 3D Self-Supervised Learning](<https://devfeed.tech/articles/uncovering-the-invisible-in-routine-radiology-cross-modal-detection-of-mri-verified-soft-tissue-pathologies-from-cbct-via-physics-guided-3d-self-supervised-learning-141834.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.23042>)

Author: Evgeny Alves Limarenko

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Medical imaging](<https://devfeed.tech/topics/medical-imaging.md>), [Physics-guided deep learning](<https://devfeed.tech/topics/physics-guided-deep-learning.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Self-supervised learning](<https://devfeed.tech/topics/self-supervised-learning.md>), [Representation learning](<https://devfeed.tech/topics/representation-learning.md>)

Tags: [3d](<https://devfeed.tech/tags/3d.md>), [architectures](<https://devfeed.tech/tags/architectures.md>), [autoencoder](<https://devfeed.tech/tags/autoencoder.md>), [automated](<https://devfeed.tech/tags/automated.md>), [beam](<https://devfeed.tech/tags/beam.md>), [consistency](<https://devfeed.tech/tags/consistency.md>), [cross](<https://devfeed.tech/tags/cross.md>), [cross-validation](<https://devfeed.tech/tags/cross-validation.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [data](<https://devfeed.tech/tags/data.md>), [deep-learning](<https://devfeed.tech/tags/deep-learning.md>), [detection](<https://devfeed.tech/tags/detection.md>), [disease](<https://devfeed.tech/tags/disease.md>), [eess-iv](<https://devfeed.tech/tags/eess-iv.md>), [flow](<https://devfeed.tech/tags/flow.md>), [framework](<https://devfeed.tech/tags/framework.md>), [gradient](<https://devfeed.tech/tags/gradient.md>), [learning](<https://devfeed.tech/tags/learning.md>), [modal](<https://devfeed.tech/tags/modal.md>), [physics](<https://devfeed.tech/tags/physics.md>), [routine](<https://devfeed.tech/tags/routine.md>), [self-supervised-learning](<https://devfeed.tech/tags/self-supervised-learning.md>)

### AI overview

In 212 paired TMJ CBCT-MRI joints, a physics-guided 3D learning pipeline detected a sub-visual signal for MRI-verified synovial effusion (ROC AUC 0.793). It used a frozen masked autoencoder pre-trained on 6,550 unlabeled CBCT volumes, with PhysLoss and the GCond optimizer directing learning toward the −300 to 300 HU soft-tissue window, followed by automated localization and cropping that removed 81% of background osseous variance. Under patient-isolated cross-validation, only representations trained with PhysLoss and GCond captured anatomical chirality and intra-patient consistency; the same architectures using standard L2 loss or MVC performed at random-guessing levels. The findings are limited to this TMJ study and support potential opportunistic screening from routine CBCT, not replacement of MRI.

### Source excerpt

arXiv:2611.23042v1 Announce Type: cross Abstract: Routine X-ray modalities like cone-beam computed tomography (CBCT) are optimized for high-contrast mineralized tissues, leaving soft-tissue abnormalities buried beneath photon noise and below human visual perception. Thus, detecting non-calcified pathologies requires secondary magnetic resonance imaging (MRI). We demonstrate that sub-visual radiographic signatures of soft-tissue pathology are physically preserved within routine CBCT scans and can be extracted for opportunistic screening using a targeted, physics-guided deep learning pipeline, even under extreme paired-data scarcity. Using the temporomandibular joint (TMJ) as a model system (N = 212 paired CBCT-MRI joints), we address the severe overfitting of high-capacity 3D networks through a two-stage framework. First, we employ a frozen full-grid 3D masked autoencoder (BertsWin) pre-trained on 6,550 unlabeled CBCT volumes via the GradientConductor (GCond) optimizer and an anatomy-aware objective (PhysLoss) redirecting gradient flow into the [-300, 300] HU soft-tissue window. Second, automated localization of the condylar zenith drives a deterministic 3D cropping that eliminates 81% of background osseous variance. Evaluated via a patient-isolated 3x10-fold stratified cross-validation protocol, a statistically robust sub-visual signal for MRI-verified synovial effusion (ROC AUC = 0.793) emerges exclusively from representations pre-trained with the PhysLoss and GCond regime. Crucially, only this setup captures true 3D anatomical chirality (p < 0.001) and high intra-patient structural consistency (0.791) over inter-patient noise; conversely, identical architectures pre-trained with standard L2 loss or unmasked variance decomposition (MVC) collapse to random guessing. These findings establish a generalizable paradigm for repurposing routine, low-contrast radiographic scans into opportunistic screening tools for occult soft-tissue disease.

## Learning Autoregressive Neural Operators on Solver-Defined Physical Constraint Manifolds

DevFeed: [Learning Autoregressive Neural Operators on Solver-Defined Physical Constraint Manifolds](<https://devfeed.tech/articles/learning-autoregressive-neural-operators-on-solver-defined-physical-constraint-manifolds-137492.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.19753>)

Author: Nannan Zhou, Guanmeng Xian, Ning Yang

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Physics-guided deep learning](<https://devfeed.tech/topics/physics-guided-deep-learning.md>), [Computational Fluid Dynamics (CFD)](<https://devfeed.tech/topics/cfd.md>)

Tags: [autoregressive](<https://devfeed.tech/tags/autoregressive.md>), [boundary](<https://devfeed.tech/tags/boundary.md>), [constraint](<https://devfeed.tech/tags/constraint.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [flow](<https://devfeed.tech/tags/flow.md>), [learning](<https://devfeed.tech/tags/learning.md>), [neural](<https://devfeed.tech/tags/neural.md>), [operator-learning](<https://devfeed.tech/tags/operator-learning.md>), [operators](<https://devfeed.tech/tags/operators.md>), [pcm](<https://devfeed.tech/tags/pcm.md>), [physics-flu-dyn](<https://devfeed.tech/tags/physics-flu-dyn.md>), [projection](<https://devfeed.tech/tags/projection.md>), [stat-ml](<https://devfeed.tech/tags/stat-ml.md>)

### AI overview

PCM-NO keeps autoregressive incompressible-flow predictions close to solver-defined feasible states by projecting each predicted increment onto admissible directions and retracting the advanced state onto the prescribed constraint set. With exact affine constraint solves, transitions remain feasible; with approximate solves, per-step retraction resets feasibility to the residual achieved by that solve, while tangent-only residuals can accumulate under an upper bound. The abstract reports experiments on periodic Navier-Stokes and lid-driven cavity flows, where PCM-NO better preserved solver-level constraints over long rollouts and under parameter, boundary-value, and resolution shifts, with lower rollout errors and better flow-specific diagnostics than the compared approaches. The supplied text is an abstract, so implementation details and comparison baselines are not available.

### Source excerpt

arXiv:2611.19753v1 Announce Type: cross Abstract: Autoregressive neural operators approximate time-dependent PDE solvers by repeatedly composing a learned transition, although the resulting states need not remain feasible under the discrete constraints of the numerical problem. We study this issue for incompressible flow, where each velocity state must satisfy solver-level incompressibility together with prescribed boundary values. We formulate autoregressive operator learning as a map between solver-defined affine feasible sets and introduce PCM-NO, a projected-residual realization of this feasible-transition principle. The neural backbone predicts a candidate state increment, a fixed tangent projection extracts its admissible component, and a metric retraction maps the advanced state to the prescribed feasible set. With exact affine constraint solves, the resulting transition is closed on the feasible set, while an affine identity separates admissible-update parameterization from state-level closure. With approximate solves, tangent-only residuals admit a cumulative upper bound, whereas per-step retraction resets feasibility to the residual achieved by the current solve. Experiments on periodic Navier--Stokes flow and lid-driven cavity flow demonstrate the superiority of PCM-NO in preserving solver-level constraints over long autoregressive rollouts and under parameter, boundary-value, and resolution shifts, while achieving lower rollout errors and better flow-specific diagnostics across both in-distribution and stress settings.

## DNA: A Double-Helix Hierarchical Deep Interaction Network Architecture for Industrial Digital Twin

DevFeed: [DNA: A Double-Helix Hierarchical Deep Interaction Network Architecture for Industrial Digital Twin](<https://devfeed.tech/articles/dna-a-double-helix-hierarchical-deep-interaction-network-architecture-for-industrial-digital-twin-139073.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.20607>)

Author: Haoran Yang, Yanan Xu, Yinan Zhang, Wenjie Zhang, Peiyu Liu, Jiahui Xiang, Wenhai Wang

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Physics-guided deep learning](<https://devfeed.tech/topics/physics-guided-deep-learning.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Deep neural networks](<https://devfeed.tech/topics/deep-neural-networks.md>)

Tags: [architecture](<https://devfeed.tech/tags/architecture.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [deep](<https://devfeed.tech/tags/deep.md>), [deep-learning](<https://devfeed.tech/tags/deep-learning.md>), [digital](<https://devfeed.tech/tags/digital.md>), [digital-twin](<https://devfeed.tech/tags/digital-twin.md>), [dna](<https://devfeed.tech/tags/dna.md>), [helix](<https://devfeed.tech/tags/helix.md>), [hybrid](<https://devfeed.tech/tags/hybrid.md>), [industrial](<https://devfeed.tech/tags/industrial.md>), [interaction](<https://devfeed.tech/tags/interaction.md>), [modeling](<https://devfeed.tech/tags/modeling.md>), [models](<https://devfeed.tech/tags/models.md>), [network](<https://devfeed.tech/tags/network.md>), [network-architecture](<https://devfeed.tech/tags/network-architecture.md>), [novel](<https://devfeed.tech/tags/novel.md>), [structure](<https://devfeed.tech/tags/structure.md>), [systems](<https://devfeed.tech/tags/systems.md>)

### AI overview

DNA combines mechanistic and data-driven modeling through two interaction strategies, then uses a hierarchical, multi-stage structure implemented with two tailored deep learning frameworks to model complex industrial systems. Evaluated on two industrial systems, it reduced MAE by 3.704% and 3.175%, respectively, compared with the strongest baseline. The supplied abstract does not describe the interaction strategies or frameworks in detail.

### Source excerpt

arXiv:2611.20607v1 Announce Type: new Abstract: High-fidelity modeling is fundamental to industrial digital twin, yet a trade-off persists between the physical transparency of mechanistic models and the superior predictive power of data-driven approaches. Existing hybrid methods often fail to bridge this gap, suffering from shallow integration and rigid one-shot modeling strategies that struggle with complex industrial systems. To address these issues, we propose a \textbf{Double-Helix Hierarchical Deep Interaction Network Architecture (DNA)}. Inspired by the double-helix structure of biological \textit{DNA}, our architecture intertwines mechanistic and data-driven modeling strands through two novel interaction strategies, forming a symbiotic architecture in which the two strands mutually enhance predictive capability. Furthermore, the hierarchical structure of DNA allows it to model complex systems in a multi-stage manner through two tailored deep learning frameworks. We evaluate the proposed approach on two distinct industrial systems, where DNA achieves reductions in MAE of 3.704\% and 3.175\%, respectively, compared to the strongest baseline. These results demonstrate the effectiveness of DNA and suggest its potential to provide new insights for the development of industrial digital twins.

## Score-Based Stochastic Closure Modeling via Offline Indirect Learning and Online Data Assimilation

DevFeed: [Score-Based Stochastic Closure Modeling via Offline Indirect Learning and Online Data Assimilation](<https://devfeed.tech/articles/score-based-stochastic-closure-modeling-via-offline-indirect-learning-and-online-data-assimilation-138907.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.19676>)

Author: Xinghao Dong, Huchen Yang, Jin-Long Wu

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Physics-guided deep learning](<https://devfeed.tech/topics/physics-guided-deep-learning.md>)

Tags: [additive-manufacturing](<https://devfeed.tech/tags/additive-manufacturing.md>), [ai](<https://devfeed.tech/tags/ai.md>), [closure](<https://devfeed.tech/tags/closure.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [data-driven](<https://devfeed.tech/tags/data-driven.md>), [diffusion](<https://devfeed.tech/tags/diffusion.md>), [diffusion-models](<https://devfeed.tech/tags/diffusion-models.md>), [digital-twins](<https://devfeed.tech/tags/digital-twins.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [gaussian](<https://devfeed.tech/tags/gaussian.md>), [learning](<https://devfeed.tech/tags/learning.md>), [modeling](<https://devfeed.tech/tags/modeling.md>), [offline](<https://devfeed.tech/tags/offline.md>), [online](<https://devfeed.tech/tags/online.md>), [physics-comp-ph](<https://devfeed.tech/tags/physics-comp-ph.md>), [physics-flu-dyn](<https://devfeed.tech/tags/physics-flu-dyn.md>), [score](<https://devfeed.tech/tags/score.md>), [simulations](<https://devfeed.tech/tags/simulations.md>), [state](<https://devfeed.tech/tags/state.md>), [statistics](<https://devfeed.tech/tags/statistics.md>), [system](<https://devfeed.tech/tags/system.md>)

### AI overview

The framework uses score-based diffusion models to learn stochastic closures without direct closure data, combining offline indirect training with online data assimilation. In experiments on two-dimensional turbulent flows, indirect learning captured key closure statistics and improved system-state predictions. Pairing the learned closure with known physical dynamics generalized across regimes better than models that directly predict state. Posterior-score assimilation represented non-Gaussian closure features better than standard Gaussian-assumption assimilation; adding state-level assimilation further improved predictions. The evidence is limited to the abstract and reported experiments.

### Source excerpt

arXiv:2611.19676v1 Announce Type: new Abstract: Stochastic closure modeling is an emerging area in reduced-order modeling of large and complex (e.g., multiscale, multi-physics) systems, for which the computationally affordable reduced-order physical descriptions (e.g., digital twins for additive manufacturing, world models for AI agents) in numerical simulations inevitably lead to information losses that cannot be well addressed by classical deterministic closures. However, learning stochastic closures typically requires closure data that are difficult to obtain from measurements or often prohibitively expensive to compute using first-principles simulations. In this work, we develop a stochastic closure modeling framework that builds on score-based diffusion models, leveraging indirect offline training and online data assimilation without requiring the direct data of closures. Numerical experiments on two-dimensional turbulent flows demonstrate that indirect offline learning captures key closure statistics and improves predictions of system state. In addition, combining the learned closure with known physical dynamics leads to better generalization capability across different regimes than purely data-driven models that directly predict system state. Compared to standard data assimilation methods with a Gaussian assumption, the posterior-score assimilation better represents non-Gaussian features of the learned closure, and its combination with state-level data assimilation further improves prediction of the system state. The proposed framework enables stochastic closure modeling in realistic and challenging scenarios of complex dynamical systems in mechanics and engineering problems, where non-Gaussian statistics of model-form uncertainties are non-negligible (e.g., due to the nonlinearities of the system dynamics and/or the observations) and true closure data are not available, for both offline learning and online assimilation.

## A Variational Latent Equilibrium for Learning in Neuronal Circuits

DevFeed: [A Variational Latent Equilibrium for Learning in Neuronal Circuits](<https://devfeed.tech/articles/a-variational-latent-equilibrium-for-learning-in-neuronal-circuits-134967.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2603.09600>)

Author: Simon Brandt, Paul Haider, Walter Senn, Federico Benitez, Mihai A. Petrovici

Published: 2026-11-16T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Neural Network](<https://devfeed.tech/topics/neural-network.md>), [Physics-guided deep learning](<https://devfeed.tech/topics/physics-guided-deep-learning.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [approximation](<https://devfeed.tech/tags/approximation.md>), [backpropagation](<https://devfeed.tech/tags/backpropagation.md>), [brain](<https://devfeed.tech/tags/brain.md>), [circuits](<https://devfeed.tech/tags/circuits.md>), [credit-assignment](<https://devfeed.tech/tags/credit-assignment.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ne](<https://devfeed.tech/tags/cs-ne.md>), [cs-sy](<https://devfeed.tech/tags/cs-sy.md>), [deep-learning](<https://devfeed.tech/tags/deep-learning.md>), [eess-sy](<https://devfeed.tech/tags/eess-sy.md>), [energy](<https://devfeed.tech/tags/energy.md>), [error](<https://devfeed.tech/tags/error.md>), [framework](<https://devfeed.tech/tags/framework.md>), [learning](<https://devfeed.tech/tags/learning.md>), [learning-algorithms](<https://devfeed.tech/tags/learning-algorithms.md>), [local](<https://devfeed.tech/tags/local.md>), [networks](<https://devfeed.tech/tags/networks.md>), [neuron](<https://devfeed.tech/tags/neuron.md>), [patterns](<https://devfeed.tech/tags/patterns.md>), [physics-bio-ph](<https://devfeed.tech/tags/physics-bio-ph.md>), [plausible](<https://devfeed.tech/tags/plausible.md>), [principles](<https://devfeed.tech/tags/principles.md>), [q-bio-nc](<https://devfeed.tech/tags/q-bio-nc.md>), [real-time](<https://devfeed.tech/tags/real-time.md>), [theory](<https://devfeed.tech/tags/theory.md>)

### AI overview

The paper proposes a time-continuous, biologically plausible approximation to backpropagation through time for neuronal networks. Starting from a prospective energy function of neuronal states, it derives adjoint-method error dynamics. Because temporal locality requires approximations, the authors identify a gain error in Generalized Latent Equilibrium (GLE) that can impair processing complex signals. They add a learning rule for backward weights to correct that difference, yielding neuron and synapse dynamics that are local in space and time. The abstract presents this as a theoretical framework and blueprint for physical circuits; it does not report empirical validation.

### Source excerpt

arXiv:2603.09600v3 Announce Type: replace-cross Abstract: Brains remain unrivaled in their ability to recognize and generate complex spatiotemporal patterns. While AI is able to reproduce some of these capabilities, deep learning algorithms remain largely at odds with our current understanding of brain circuitry and dynamics. This is prominently the case for backpropagation through time (BPTT), the go-to algorithm for learning complex temporal dependencies. In this work we propose a general formalism to approximate BPTT in a controlled, biologically plausible manner. Our approach builds on, unifies and extends several previous approaches to local, time-continuous, phase-free spatiotemporal credit assignment based on principles of energy conservation and extremal action. Our starting point is a prospective energy function of neuronal states, from which we calculate real-time error dynamics for time-continuous neuronal networks. With this, as a first contribution we provide a simple and straightforward derivation of the adjoint method dynamical equations, the time-continuous equivalent to BPTT. Importantly, additional approximations are necessary to avoid unphysical temporal non-localities. One such approximation was recently proposed in Generalized Latent Equilibrium (GLE), but enforcing temporal locality of the backpropagating dynamics induces an error in the gain that becomes problematic when processing complex signals. Thus, our second contribution is to design a learning rule for the backward weights to find a fully local (in space and time) set of equations for neuron and synapse dynamics that corrects for this gain difference. With this inclusion of backward weight learning, our theory provides a rigorous framework for spatiotemporal deep learning in the brain, while simultaneously advancing a blueprint for physical circuits capable of carrying out these computations.

## A mesh-free deep energy method for phase-field modeling of thermally driven brittle fracture

DevFeed: [A mesh-free deep energy method for phase-field modeling of thermally driven brittle fracture](<https://devfeed.tech/articles/a-mesh-free-deep-energy-method-for-phase-field-modeling-of-thermally-driven-brittle-fracture-135812.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.17630>)

Author: Han Zhang, Mehrisadat Makki Alamdari, Babak Shahbodagh, Mohammad Vahab, Cosmin Anitescu, Timon Rabczuk, Elena Atroshchenko

Published: 2026-11-16T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Deep learning](<https://devfeed.tech/topics/deep-learning.md>), [Physics-guided deep learning](<https://devfeed.tech/topics/physics-guided-deep-learning.md>), [Simulation](<https://devfeed.tech/topics/simulation.md>), [Finite Element Method (FEM)](<https://devfeed.tech/topics/finite-element-method.md>)

Tags: [array](<https://devfeed.tech/tags/array.md>), [arrays](<https://devfeed.tech/tags/arrays.md>), [arrest](<https://devfeed.tech/tags/arrest.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [ceramic](<https://devfeed.tech/tags/ceramic.md>), [cond-mat-mtrl-sci](<https://devfeed.tech/tags/cond-mat-mtrl-sci.md>), [crack](<https://devfeed.tech/tags/crack.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [deep](<https://devfeed.tech/tags/deep.md>), [encoding](<https://devfeed.tech/tags/encoding.md>), [experiment](<https://devfeed.tech/tags/experiment.md>), [fields](<https://devfeed.tech/tags/fields.md>), [free](<https://devfeed.tech/tags/free.md>), [functional](<https://devfeed.tech/tags/functional.md>), [heat-equation](<https://devfeed.tech/tags/heat-equation.md>), [loading](<https://devfeed.tech/tags/loading.md>), [modeling](<https://devfeed.tech/tags/modeling.md>), [neural](<https://devfeed.tech/tags/neural.md>), [prediction](<https://devfeed.tech/tags/prediction.md>), [reference](<https://devfeed.tech/tags/reference.md>), [spacing](<https://devfeed.tech/tags/spacing.md>), [temperature](<https://devfeed.tech/tags/temperature.md>)

### AI overview

The proposed mesh-free deep energy method predicts spontaneous crack formation under transient thermal loading without prescribed crack paths, nucleation sites, or crack-adapted meshes. Separate multiresolution networks represent displacement and phase fields and temperature; energy terms are estimated at integration points redrawn each iteration. Restricting irreversibility to the crack set, carrying optimizer state across load steps, and decaying learning rates within steps help limit spurious damage. Reported results include initiation peak loads on two notched plates that closely match an independent finite element reference, and crack patterns in glass and ceramic cases consistent with stability predictions or published observations and benchmarks. However, crack arrays in a quenched slab and annulus are denser than finite element results on the same material field.

### Source excerpt

arXiv:2611.17630v1 Announce Type: new Abstract: Phase-field simulation of thermally driven fracture requires resolving localized crack bands whose paths are unknown in advance and thermal boundary layers whose thickness evolves during the transient. Neural energy-based computations of thermal fracture have so far started from defects placed in advance. This paper proposes a mesh-free deep energy method based on multiresolution neural representations that predicts spontaneous crack nucleation, crack spacing, and selective crack propagation and arrest under transient thermal loading. Two networks, each with its own multiresolution encoding, represent the displacement and phase fields and the transient temperature, and the mechanical energy and the incremental functional of the backward Euler heat equation are estimated on integration points redrawn at every iteration. Restricting irreversibility to the crack set, carrying the optimizer state across load steps and decaying the learning rates within each step help prevent spurious damage accumulation. On two notched plates the initiation peak loads agree closely with an independent finite element reference. On a quenched glass strip under an analytical temperature field the crack path changes from straight to oscillatory and then to branched, with an oscillation threshold close to the linear stability prediction. Under a prescribed boundary temperature a quenched ceramic slab forms a crack array of the type observed in experiment, with a spacing close to that of a published benchmark. On the slab and a quenched annulus the arrays are denser than those of a finite element computation on the same material field. No crack paths, nucleation sites or crack-adapted meshes are prescribed.

## Retrieval of Coastal Biogeochemical Parameters From Near-Surface Hyperspectral Remote Sensing Reflectance Using Physics-Aware Meta-Learning

DevFeed: [Retrieval of Coastal Biogeochemical Parameters From Near-Surface Hyperspectral Remote Sensing Reflectance Using Physics-Aware Meta-Learning](<https://devfeed.tech/articles/retrieval-of-coastal-biogeochemical-parameters-from-near-surface-hyperspectral-remote-sensing-reflectance-using-physics-aware-meta-learning-135683.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2605.05623>)

Author: Yiqing Guo, Nagur R. C. Cherukuru, Eric A. Lehmann, S. L. Kesav Unnithan, Tim J. Malthus, Gemma Kerrisk, Xiubin Qi, Faisal Islam, Tisham Dhar, Mark J. Doubell

Published: 2026-11-16T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Physics-guided deep learning](<https://devfeed.tech/topics/physics-guided-deep-learning.md>), [Training AI Models](<https://devfeed.tech/topics/training-ai-models.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Fine-tuning](<https://devfeed.tech/topics/fine-tuning.md>)

Tags: [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [learning](<https://devfeed.tech/tags/learning.md>), [meta](<https://devfeed.tech/tags/meta.md>), [physics](<https://devfeed.tech/tags/physics.md>), [remote-sensing](<https://devfeed.tech/tags/remote-sensing.md>), [retrieval](<https://devfeed.tech/tags/retrieval.md>), [water-quality](<https://devfeed.tech/tags/water-quality.md>)

### AI overview

The framework aims to improve coastal water-quality retrieval across regions, where reflectance-parameter relationships vary with environmental conditions and biogeochemistry. It first meta-trains a region-agnostic model on synthetic reflectance and water-quality samples generated by a bio-optical forward model using an in situ spectral library, then fine-tunes the model on local samples. The study evaluates this approach using in situ hyperspectral reflectance and water-quality measurements from five geographically distinct Australian coastal sites. The excerpt does not report performance results.

### Source excerpt

arXiv:2605.05623v4 Announce Type: replace Abstract: Hyperspectral in situ sensing has shown promise in retrieving aquatic water quality (WQ) parameters, such as total suspended solids (TSS), dissolved organic carbon (DOC), and total chlorophyll-a (TChl-a), for cost-effective management of coastal WQ. However, generalising such retrieval algorithms across water bodies remains challenging, as the relationship between remote sensing reflectance ($R_{rs}$) and WQ parameters can vary considerably from one region to another due to regional distinctions in environmental conditions and biogeochemistry that lead to different WQ ranges and bio-optical properties. In this study, we propose a two-stage physics-aware meta-learning framework for retrieving coastal WQ parameters from near-surface $R_{rs}$ observations. Unlike purely data-driven approaches trained solely on paired observations, the proposed framework incorporates physical knowledge through a two-stage learning process. In the first stage, a region-agnostic base model is pretrained using meta-learning on synthetic WQ--$R_{rs}$ samples generated by a bio-optical forward model parameterised with an in situ bio-optical spectral library. In the second stage, the pretrained model is fine-tuned using local samples to adapt to regional conditions. To evaluate the proposed approach, we collected in situ hyperspectral $R_{rs}$ and WQ measurements from five geographically distinct sites in Australian coastal waters.

## CARE-Drive: Coupled Alternating Refinement for Ego-trajectories

DevFeed: [CARE-Drive: Coupled Alternating Refinement for Ego-trajectories](<https://devfeed.tech/articles/care-drive-coupled-alternating-refinement-for-ego-trajectories-132125.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.15360>)

Author: Xiao Ma, Yijun Chen, Zhiquan Hu, Jicheng Zhao, Yuming Li, Chuang Dai

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [autonomous vehicles](<https://devfeed.tech/topics/autonomous-vehicles.md>), [Enhancing AI model performance](<https://devfeed.tech/topics/enhancing-ai-model-performance.md>), [Fine-tuning](<https://devfeed.tech/topics/fine-tuning.md>), [Physics-guided deep learning](<https://devfeed.tech/topics/physics-guided-deep-learning.md>)

Tags: [ablation](<https://devfeed.tech/tags/ablation.md>), [angular](<https://devfeed.tech/tags/angular.md>), [autonomous](<https://devfeed.tech/tags/autonomous.md>), [autonomous-driving](<https://devfeed.tech/tags/autonomous-driving.md>), [coupling](<https://devfeed.tech/tags/coupling.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [denoising](<https://devfeed.tech/tags/denoising.md>), [driving](<https://devfeed.tech/tags/driving.md>), [end-to-end-autonomous-driving](<https://devfeed.tech/tags/end-to-end-autonomous-driving.md>), [experts](<https://devfeed.tech/tags/experts.md>), [fine-tuning](<https://devfeed.tech/tags/fine-tuning.md>), [imitation-learning](<https://devfeed.tech/tags/imitation-learning.md>), [knowledge](<https://devfeed.tech/tags/knowledge.md>), [metric](<https://devfeed.tech/tags/metric.md>), [multimodal](<https://devfeed.tech/tags/multimodal.md>), [optimization](<https://devfeed.tech/tags/optimization.md>), [orientation](<https://devfeed.tech/tags/orientation.md>), [position](<https://devfeed.tech/tags/position.md>), [prediction](<https://devfeed.tech/tags/prediction.md>), [vision-language-action](<https://devfeed.tech/tags/vision-language-action.md>), [vla](<https://devfeed.tech/tags/vla.md>)

### AI overview

CARE-Drive refines driving trajectories by separating waypoint position correction from heading generation while keeping the two conditionally linked. A frozen backbone supplies predictive representations and a base trajectory; a position expert predicts bounded metric residuals, and a heading expert reconstructs vehicle orientation from adjacent angular increments along the corrected path. The experts are alternately optimized after supervised initialization, with reinforcement fine-tuning. On NAVSIM v1, the method reports 88.8 PDMS, 2.4 points above the same backbone's unrefined trajectories; ablations support the specialized experts and staged training.

### Source excerpt

arXiv:2611.15360v1 Announce Type: cross Abstract: Vision-language-action (VLA) models bring multimodal understanding and world knowledge to end-to-end autonomous driving. To further improve the quality of predicted trajectories, existing studies introduce candidate refinement, residual denoising, and token correction. However, these methods typically focus on unified refinement of the entire trajectory, paying limited attention to the distinct physical quantities and their interactions. We further investigate physical specialization and coordination in trajectory refinement, centering on position and heading for simpler and more intuitive correction modeling. To this end, we propose CARE-Drive (Coupled Alternating Refinement for Ego-trajectories), which decomposes trajectory refinement into waypoint residual correction and position-conditioned heading generation. Through imitation learning and a future-token auxiliary task, the backbone provides predictive representations and a base trajectory. The position expert predicts bounded metric waypoint residuals, while the heading expert reconstructs vehicle orientation from adjacent angular increments conditioned on the corrected path. With the backbone frozen, the two experts are optimized alternately through supervised initialization and reinforcement fine-tuning. This design simplifies refinement through clear physical specialization while preserving conditional coupling during prediction and optimization. On NAVSIM v1, CARE-Drive achieves 88.8 PDMS, improving by 2.4 points over the unrefined base trajectories generated by the same backbone. Ablation studies further support the effectiveness of specialized experts and staged training.

## A physics-constrained deep reinforcement learning method for autonomous mission planning of Titan deep space probes

DevFeed: [A physics-constrained deep reinforcement learning method for autonomous mission planning of Titan deep space probes](<https://devfeed.tech/articles/a-physics-constrained-deep-reinforcement-learning-method-for-autonomous-mission-planning-of-titan-deep-space-probes-132126.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.15389>)

Author: Hanmin Li, Tao Wang, Shuangliang Liu, Zhe Zhang, Maodeng Li

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Reinforcement learning](<https://devfeed.tech/topics/reinforcement-learning.md>), [Physics-guided deep learning](<https://devfeed.tech/topics/physics-guided-deep-learning.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>)

Tags: [ablation](<https://devfeed.tech/tags/ablation.md>), [autonomous](<https://devfeed.tech/tags/autonomous.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [deep](<https://devfeed.tech/tags/deep.md>), [deep-reinforcement-learning](<https://devfeed.tech/tags/deep-reinforcement-learning.md>), [learning](<https://devfeed.tech/tags/learning.md>), [mission](<https://devfeed.tech/tags/mission.md>), [physics](<https://devfeed.tech/tags/physics.md>), [planning](<https://devfeed.tech/tags/planning.md>), [reinforcement-learning](<https://devfeed.tech/tags/reinforcement-learning.md>), [space](<https://devfeed.tech/tags/space.md>), [titan](<https://devfeed.tech/tags/titan.md>)

### AI overview

The paper proposes PCM-PPO for autonomous mission planning of Titan probes, where long communication delays make continuous ground-in-the-loop planning impractical. It models mission feasibility and resources in a Titan-specific discrete-time Markov decision process, with dynamics that include central gravity, J2 perturbation, third-body attraction, and downlink occultation. A fixed-capacity target pool handles predefined and opportunity targets; the mission state combines target-wise and global representations. A physics-constrained action mask limits policy choices to decisions admissible under the current mission state, steering learning away from infeasible actions. The authors report comparative, ablation, and uncertainty experiments validating the method, and describe it as feasible and interpretable. The supplied material is an abstract and does not provide experiment details or quantitative results.

### Source excerpt

arXiv:2611.15389v1 Announce Type: new Abstract: Titan is a compelling target for future deep space exploration, while its extreme Earth-Titan distance and complex environment make continuous ground-in-the-loop planning impractical. Existing planning methods, primarily developed for Earth-observation scenarios, have limited generalizability to uncertain deep space conditions and adaptability to time-varying mission states. To address these challenges, this paper proposes a physics-constrained maskable proximal policy optimization (PCM-PPO) method for Titan autonomous mission planning. A Titan-specific discrete-time Markov decision process is established to characterize mission feasibility and resource constraints. A system dynamics and geometric model is incorporated to enhance physical representation, including central gravity, J2 perturbation, third-body attraction, and downlink occultation. To accommodate opportunity targets under a bounded action dimension, PCM-PPO organizes predefined and opportunity targets through a fixed-capacity target pool and represents the mission state with a target-wise state matrix and a global state vector. A physics-constrained action mask constructs an admissible decision domain from current mission states, guiding policy learning toward physically executable decisions and avoiding infeasible exploration. An integrated training strategy facilitates policy optimization under coupled constraints. Comparative, ablation, and uncertainty experiments validate the effectiveness of PCM-PPO. PCM-PPO provides a feasible and interpretable scheme for autonomous mission planning of Titan deep space probes.

## Learning Temporal Dynamics with Derivative Representations for Time Series Forecasting

DevFeed: [Learning Temporal Dynamics with Derivative Representations for Time Series Forecasting](<https://devfeed.tech/articles/learning-temporal-dynamics-with-derivative-representations-for-time-series-forecasting-133123.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.16514>)

Author: Shujie Duan, Jia Hong Puah, Haocheng Wu, Yifei Mao, Hongting Zhou, Chenxuan Li, Zitao Fang, Yang Lu, Sim Kuan Goh

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Time Series](<https://devfeed.tech/topics/time-series.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Temporal data](<https://devfeed.tech/topics/temporal-data.md>), [Physics-guided deep learning](<https://devfeed.tech/topics/physics-guided-deep-learning.md>)

Tags: [coherence](<https://devfeed.tech/tags/coherence.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [deep-learning](<https://devfeed.tech/tags/deep-learning.md>), [forecasting](<https://devfeed.tech/tags/forecasting.md>), [learning](<https://devfeed.tech/tags/learning.md>), [model](<https://devfeed.tech/tags/model.md>), [model-agnostic](<https://devfeed.tech/tags/model-agnostic.md>), [series](<https://devfeed.tech/tags/series.md>), [temporal](<https://devfeed.tech/tags/temporal.md>), [time](<https://devfeed.tech/tags/time.md>), [time-series](<https://devfeed.tech/tags/time-series.md>), [time-series-forecasting](<https://devfeed.tech/tags/time-series-forecasting.md>), [training](<https://devfeed.tech/tags/training.md>), [transformers](<https://devfeed.tech/tags/transformers.md>)

### AI overview

DeltaTSF trains forecasting models to jointly represent time series and their discrete temporal derivatives across input and forecast horizons. Coupling these representations is intended to encourage temporal consistency, described as a form of discrete Taylor consistency, and produce more stable, dynamically coherent forecasts. The framework is model-agnostic, requires no explicit derivative computation, and adds no inference-time computational overhead. The abstract reports consistent performance improvements and more coherent predictions across diverse benchmarks and backbone architectures, but provides no specific results or benchmark details. Source code is stated to be forthcoming upon acceptance.

### Source excerpt

arXiv:2611.16514v1 Announce Type: new Abstract: Time series forecasting aims to predict the future evolution of complex systems by learning their underlying dynamics from past observations. Despite recent advances in deep learning, Transformers in particular still struggle to capture the temporal evolution of system dynamics. Existing approaches have been largely driven by architectural innovation, with comparatively little attention paid to training strategies that explicitly exploit the temporal structure governing system dynamics. In this work, we introduce DeltaTSF, a dynamics-aware forecasting framework based on discrete derivative representations of time series. Specifically, DeltaTSF jointly learns the series and their discrete temporal derivatives across both input and forecast horizons within a unified, model-agnostic framework. By coupling these representations over time, the model is encouraged to maintain temporal consistency, a property we show to induce a form of discrete Taylor consistency during training. This formulation enables the model to better reflect the underlying system dynamics, resulting in more stable and physically coherent forecasts. Notably, DeltaTSF does not require explicit derivative computation and introduces no additional computational overhead at inference time. Extensive experiments across diverse benchmarks and backbone architectures demonstrate that DeltaTSF consistently improves forecasting performance and yields more dynamically coherent predictions than strong SOTA methods. These results highlight the effectiveness of enforcing dynamic coherence through derivative representations as a model-agnostic paradigm that operates alongside and complements architectural design. Source code will be made publicly available upon acceptance.

## Block-Sparse Gauss-Newton Optimization of Finite Basis Physics-Informed Neural Networks: Sparse Automatic Differentiation Approach

DevFeed: [Block-Sparse Gauss-Newton Optimization of Finite Basis Physics-Informed Neural Networks: Sparse Automatic Differentiation Approach](<https://devfeed.tech/articles/block-sparse-gauss-newton-optimization-of-finite-basis-physics-informed-neural-networks-sparse-automatic-differentiation-approach-128678.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.13231>)

Author: Denis Korolev, Adrian Hill, Jeeeun Lee

Published: 2026-11-11T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Physics-guided deep learning](<https://devfeed.tech/topics/physics-guided-deep-learning.md>), [Neural Network](<https://devfeed.tech/topics/neural-network.md>), [Operations research and optimization](<https://devfeed.tech/topics/operations-research-and-optimization.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>)

Tags: [automatic](<https://devfeed.tech/tags/automatic.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cs-na](<https://devfeed.tech/tags/cs-na.md>), [domain-decomposition](<https://devfeed.tech/tags/domain-decomposition.md>), [graph-coloring](<https://devfeed.tech/tags/graph-coloring.md>), [linear-algebra](<https://devfeed.tech/tags/linear-algebra.md>), [math-na](<https://devfeed.tech/tags/math-na.md>), [neural-networks](<https://devfeed.tech/tags/neural-networks.md>), [optimization](<https://devfeed.tech/tags/optimization.md>), [physics-informed](<https://devfeed.tech/tags/physics-informed.md>), [sparsity](<https://devfeed.tech/tags/sparsity.md>)

### AI overview

The paper uses FBPINN overlapping domain decomposition to exploit block sparsity in the residual Jacobian and Gauss-Newton matrix. Graph coloring groups structurally orthogonal Jacobian columns or rows for sparse automatic differentiation, reducing differentiation cost and memory while retaining structurally nonzero derivatives. The compressed Jacobian is assembled directly in block-sparse form, allowing sparse direct solvers or preconditioned iterative methods. In reported multiscale PDE experiments, the method achieves high accuracy and substantially faster optimization than first-order baselines, while remaining tractable at large problem sizes.

### Source excerpt

arXiv:2611.13231v1 Announce Type: cross Abstract: In this work, we develop an efficient Gauss-Newton optimization approach for finite basis physics-informed neural networks (FBPINNs). The key ingredient is the overlapping domain decomposition architecture of FBPINNs, which localizes the influence of each subnetwork and induces block sparsity in both the residual Jacobian and the corresponding Gauss-Newton matrix. We characterize this structure and exploit it through sparse automatic differentiation, using graph coloring to partition structurally orthogonal Jacobian columns or rows. This reduces differentiation costs and memory requirements while preserving all structurally nonzero derivative information. The resulting compressed Jacobian representations allow us to assemble the Gauss-Newton system directly in block-sparse form, enabling the use of sparse direct solvers and preconditioned iterative methods. In this way, the otherwise expensive second-order optimization problem is reduced to scalable sparse linear algebra. We demonstrate the approach on multiscale PDEs, where accurate approximation requires rich neural network parametrizations and large numbers of collocation points. Numerical experiments show that our block-sparse Gauss-Newton method attains high accuracy and substantially accelerates optimization relative to first-order baselines while remaining computationally tractable for large problem sizes.

## RamanPFN: A physics-guided spectral foundation framework for in-context learning across diverse Raman applications

DevFeed: [RamanPFN: A physics-guided spectral foundation framework for in-context learning across diverse Raman applications](<https://devfeed.tech/articles/ramanpfn-a-physics-guided-spectral-foundation-framework-for-in-context-learning-across-diverse-raman-applications-128135.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2608.02157>)

Author: Xingyu Pan, Huan Wang, Jinjia Guo, Zhenlin Zhao, Siming Dong, Jixi Lu

Published: 2026-11-11T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Physics-guided deep learning](<https://devfeed.tech/topics/physics-guided-deep-learning.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [AI Foundation Models](<https://devfeed.tech/topics/ai-foundation-models.md>), [Representation learning](<https://devfeed.tech/topics/representation-learning.md>), [few-shot](<https://devfeed.tech/topics/few-shot.md>)

Tags: [classification](<https://devfeed.tech/tags/classification.md>), [context](<https://devfeed.tech/tags/context.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [foundation](<https://devfeed.tech/tags/foundation.md>), [foundation-models](<https://devfeed.tech/tags/foundation-models.md>), [framework](<https://devfeed.tech/tags/framework.md>), [learning](<https://devfeed.tech/tags/learning.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [materials-science](<https://devfeed.tech/tags/materials-science.md>), [physics](<https://devfeed.tech/tags/physics.md>), [raman-spectroscopy](<https://devfeed.tech/tags/raman-spectroscopy.md>), [regression](<https://devfeed.tech/tags/regression.md>), [spectroscopy](<https://devfeed.tech/tags/spectroscopy.md>)

### AI overview

RamanPFN addresses a limitation of in-context tabular models on wide Raman spectra: although feature subsampling can cover every channel, it may rarely expose pairs of channels together. It combines global compositional unmixing, which links covarying peaks through shared non-negative bases, with local vibrational subspace encoding, which preserves peak shape, intensity changes, and shifts. Posterior marginalization over admissible unmixings enables prediction without model updates. Across 74 public datasets (over 330,000 spectra), it outperformed 26 methods on 129 regression targets, reducing backbone RMSE by 13.2%. Tests with TabPFN-3, TabICLv2, and Kumo-Tabular also report lower regression error (12.3%-25.7%) and multiclass log loss (42.4%-63.5%); these are abstract-reported results.

### Source excerpt

arXiv:2608.02157v3 Announce Type: replace-cross Abstract: Raman spectroscopy enables label-free molecular characterization across materials science, analytical chemistry, biomedicine, and industrial process monitoring. However, machine learning for high-dimensional spectroscopy remains constrained by limited labeled data and a mismatch between the physical organization of spectra and feature-agnostic models. Tabular foundation models such as TabPFN predict in context without parameter updates, but on wide spectra feature subsampling preserves channel coverage while losing joint visibility: in the five widest tasks, every channel was covered, yet only 4.4% of random channel pairs co-occurred in any forward pass. We present RamanPFN, a general-purpose spectral foundation framework for unified in-context inference through physics-guided spectral learning, which encodes the additive mixing of constituent spectra and the locality of vibrational bands. Global Compositional Unmixing decomposes distributed, multi-band mixture signatures into shared non-negative bases to couple covarying peaks across the full spectrum, while Local Vibrational Subspace Encoding extracts orthogonal modes within contiguous spectral neighborhoods to preserve peak morphology, intensity fluctuations, and peak shifts. Posterior marginalization over admissible unmixings unifies both scales into in-context prediction without updating the model. Across 74 public Raman datasets exceeding 330,000 spectra, RamanPFN outperformed 26 methods spanning chemometrics, spectral neural networks, deep tabular learners, and tabular foundation models on 129 regression targets, reducing backbone RMSE by 13.2%. We demonstrate across TabPFN-3, TabICLv2, and Kumo-Tabular that this advantage generalizes to both regression and classification tasks, reducing regression error by 12.3%-25.7% and multiclass log loss by 42.4%-63.5% to enable data-efficient quantitative analysis and sample identification.

## Advancing Mass-Conserving Deep Learning for Distributed Hydrologic Modeling: A Case Study in a Snow-Dominated Watershed

DevFeed: [Advancing Mass-Conserving Deep Learning for Distributed Hydrologic Modeling: A Case Study in a Snow-Dominated Watershed](<https://devfeed.tech/articles/advancing-mass-conserving-deep-learning-for-distributed-hydrologic-modeling-a-case-study-in-a-snow-dominated-watershed-128694.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.13346>)

Author: Yihan Wang, Lujun Zhang, N. Benjamin Erichson, Yang Hong, Tiantian Yang

Published: 2026-11-11T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Physics-guided deep learning](<https://devfeed.tech/topics/physics-guided-deep-learning.md>), [Deep learning](<https://devfeed.tech/topics/deep-learning.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Neural Network](<https://devfeed.tech/topics/neural-network.md>)

Tags: [architecture](<https://devfeed.tech/tags/architecture.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [case-study](<https://devfeed.tech/tags/case-study.md>), [consistency](<https://devfeed.tech/tags/consistency.md>), [constraint](<https://devfeed.tech/tags/constraint.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [deep-learning](<https://devfeed.tech/tags/deep-learning.md>), [diagnostic](<https://devfeed.tech/tags/diagnostic.md>), [diagnostics](<https://devfeed.tech/tags/diagnostics.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [dl](<https://devfeed.tech/tags/dl.md>), [framework](<https://devfeed.tech/tags/framework.md>), [gauge](<https://devfeed.tech/tags/gauge.md>), [long-short-term-memory](<https://devfeed.tech/tags/long-short-term-memory.md>), [lstm](<https://devfeed.tech/tags/lstm.md>), [model](<https://devfeed.tech/tags/model.md>), [modeling](<https://devfeed.tech/tags/modeling.md>), [models](<https://devfeed.tech/tags/models.md>), [network](<https://devfeed.tech/tags/network.md>), [physics-geo-ph](<https://devfeed.tech/tags/physics-geo-ph.md>), [watershed](<https://devfeed.tech/tags/watershed.md>)

### AI overview

PECAN is a distributed, mass-conserving convolutional LSTM framework that simulates outlet streamflow competitively while exposing spatial latent states for diagnosing hydrologic behavior. Its two variants, trained only on gauge streamflow in a snow-dominated Colorado watershed, differ in their conservation constraint: PECAN-MC requires precipitation-based water accounting, while PECAN-MCR permits additional water inputs. Benchmarked against lumped deep-learning models and the distributed VIC model, PECAN's cell states showed temporal patterns resembling reference soil moisture. However, both the reported analysis and abstract identify a snow-process failure: the model removes winter snow storage too early instead of retaining it for spring release. The authors present this as a structural limitation that streamflow accuracy alone may not reveal.

### Source excerpt

arXiv:2611.13346v1 Announce Type: new Abstract: Recent progress in physics-aware deep learning (DL) has embedded mass conservation within Long Short-Term Memory (LSTM) architecture for hydrologic modeling, improving physical consistency and enabling diagnostics of internal states. However, existing mass-conserving neural networks are primarily developed in lumped formulations. Consequently, previous studies provide only fragmented and often indirect evidence of when and why imposed physical constraints may lead to unrealistic hydrologic behavior. To address this limitation, we propose Physics-Enhanced Conservation-Augmented Net (PECAN), a mass-conserving convolutional LSTM-based framework for distributed hydrologic modeling and diagnosing structural limitations of physics-aware DL models. PECAN is implemented in two variants: (1) PECAN-MC with a strict precipitation-based conservation constraint and (2) PECAN-MCR with a relaxed conservation constraint allowing additional water inputs beyond precipitation. Both variants are trained solely on gauge streamflow and evaluated in a snow-dominated watershed in Colorado. We benchmark PECAN against lumped DL baselines and the distributed process-based Variable Infiltration Capacity (VIC) model. Results show that PECAN achieves competitive outlet streamflow simulation accuracy while producing spatially explicit latent states for interpretation. Comparisons with reference soil moisture indicate that PECAN cell states resemble meaningful temporal soil moisture dynamics. However, analyses also reveal that PECAN misrepresents snow processes by prematurely removing winter snow water storage from the system rather than retaining it for spring release. PECAN demonstrates accurate streamflow simulation at gauges and shows diagnostic potential to expose structural limitations in hydrologic process representation, thereby informing future physics-aware neural network development for hydrology.

## Physics-Constrained Autoencoder for Interpretable Health Representation Learning in Turbofan Engines

DevFeed: [Physics-Constrained Autoencoder for Interpretable Health Representation Learning in Turbofan Engines](<https://devfeed.tech/articles/physics-constrained-autoencoder-for-interpretable-health-representation-learning-in-turbofan-engines-128818.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.14266>)

Author: Leonardo Baldo, Jokin Alcibar Ibarzabal, Daniel Cisneros Acevedo, Kai Goebel, Manuel Arias Chao

Published: 2026-11-11T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Physics-guided deep learning](<https://devfeed.tech/topics/physics-guided-deep-learning.md>), [Representation learning](<https://devfeed.tech/topics/representation-learning.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>)

Tags: [autoencoder](<https://devfeed.tech/tags/autoencoder.md>), [component](<https://devfeed.tech/tags/component.md>), [compression](<https://devfeed.tech/tags/compression.md>), [condition](<https://devfeed.tech/tags/condition.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [data-driven](<https://devfeed.tech/tags/data-driven.md>), [dataset](<https://devfeed.tech/tags/dataset.md>), [decoder](<https://devfeed.tech/tags/decoder.md>), [diagnostic](<https://devfeed.tech/tags/diagnostic.md>), [eess-sp](<https://devfeed.tech/tags/eess-sp.md>), [engine](<https://devfeed.tech/tags/engine.md>), [engines](<https://devfeed.tech/tags/engines.md>), [framework](<https://devfeed.tech/tags/framework.md>), [health](<https://devfeed.tech/tags/health.md>), [inductive](<https://devfeed.tech/tags/inductive.md>), [knowledge](<https://devfeed.tech/tags/knowledge.md>), [learning](<https://devfeed.tech/tags/learning.md>), [models](<https://devfeed.tech/tags/models.md>), [monitoring](<https://devfeed.tech/tags/monitoring.md>), [operational](<https://devfeed.tech/tags/operational.md>), [physics](<https://devfeed.tech/tags/physics.md>), [turbofan](<https://devfeed.tech/tags/turbofan.md>)

### AI overview

The physics-constrained autoencoder makes latent health representations more interpretable by using differentiable physical equations in its decoder and a physics-informed loss to associate latent variables with predefined quantities. On N-CMAPSS, it achieved reconstruction R² 0.9993 and MAE 0.0036, comparable to fully data-driven models. After operating-condition correction, the inferred quantities tracked degradation; their aggregation yielded a health index with Spearman correlation 0.9791, RMSE 0.0286, and MAE 0.0233 against ground truth. A single model evaluated across four degradation scenarios retained component signatures distinguishing compression-side from expansion-side degradation without diagnostic labels. Component-map visualizations present degradation trends in a form intended to be familiar to engine operators and domain experts.

### Source excerpt

arXiv:2611.14266v1 Announce Type: new Abstract: Health Indices (HIs) summarize degradation from operational data and support diagnostic and prognostic tasks in Prognostics and Health Management (PHM). Their usefulness depends on capturing degradation and interpreting system condition. Autoencoders learn such representations from unlabelled monitoring data; however, when training is driven only by reconstruction, the physical meaning of latent variables remains undefined, limiting their interpretation and diagnostic value. This work proposes a physics-constrained autoencoder that embeds system knowledge as an inductive bias in the decoder architecture and as a learning bias in the loss function. The decoder contains differentiable physical equations that receive latent variables and operating conditions as inputs, structurally guiding each latent variable to represent a predefined physical quantity and enabling component-level analysis before aggregation into a system-level HI.The methodology is benchmarked on the N-CMAPSS dataset against alternative representations. Reconstruction reaches an $R^2$ of 0.9993 and an MAE of 0.0036, comparable to fully data-driven models. After operating-condition correction, the inferred physical quantities track degradation, and their aggregation produces an HI with a Spearman correlation of 0.9791 with the ground truth, an RMSE of 0.0286, and an MAE of 0.0233. The methodology is further evaluated using a single model trained across four datasets representing distinct degradation scenarios. This model preserves component-level signatures that distinguish compression-side from expansion-side degradation without diagnostic labels.Finally, the proposed framework enables degradation trends to be visualized on component maps, providing a health representation in a form familiar to engine operators and domain experts.

## AdjointFlow: Accelerating Scientific Inverse Problems with Physics-Based Transport

DevFeed: [AdjointFlow: Accelerating Scientific Inverse Problems with Physics-Based Transport](<https://devfeed.tech/articles/adjointflow-accelerating-scientific-inverse-problems-with-physics-based-transport-128702.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.13389>)

Author: Ethan Hanold, Nicholas M. Boffi, Qin Li, Leonardo Zepeda-N\\'u\\~nez

Published: 2026-11-11T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Physics-guided deep learning](<https://devfeed.tech/topics/physics-guided-deep-learning.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>)

Tags: [astrophysics](<https://devfeed.tech/tags/astrophysics.md>), [black-hole](<https://devfeed.tech/tags/black-hole.md>), [conditional](<https://devfeed.tech/tags/conditional.md>), [consistency](<https://devfeed.tech/tags/consistency.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [data-consistency](<https://devfeed.tech/tags/data-consistency.md>), [data-driven](<https://devfeed.tech/tags/data-driven.md>), [diffusion](<https://devfeed.tech/tags/diffusion.md>), [diffusion-models](<https://devfeed.tech/tags/diffusion-models.md>), [distribution](<https://devfeed.tech/tags/distribution.md>), [distributions](<https://devfeed.tech/tags/distributions.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [faster](<https://devfeed.tech/tags/faster.md>), [framework](<https://devfeed.tech/tags/framework.md>), [gaussian](<https://devfeed.tech/tags/gaussian.md>), [generative](<https://devfeed.tech/tags/generative.md>), [imaging](<https://devfeed.tech/tags/imaging.md>), [inference](<https://devfeed.tech/tags/inference.md>), [physics](<https://devfeed.tech/tags/physics.md>), [problems](<https://devfeed.tech/tags/problems.md>), [transport](<https://devfeed.tech/tags/transport.md>)

### AI overview

AdjointFlow incorporates known physics into generative transport for scientific inverse problems. It uses a data-dependent, adjoint-based initialization, adjusts transport velocity to remain regular when endpoint distributions are concentrated, and builds the initialization into the learned velocity network. In experiments on inverse scattering, seismic imaging, and black hole imaging, it reportedly reconstructs more accurately than conditional diffusion models and runs up to five orders of magnitude faster at inference than plug-and-play methods. Optional inference-time refinement that queries the physical model can improve data consistency and match or exceed plug-and-play accuracy without retraining.

### Source excerpt

arXiv:2611.13389v1 Announce Type: new Abstract: Inverse problems are ubiquitous across scientific applications ranging from seismic imaging to astrophysics to nondestructive testing, yet classical solvers often struggle with nonconvexity and repeated evaluations of expensive physical models. Diffusion models can amortize this computation offline, but largely data-driven conditional models underutilize known physics and suffer from ill-conditioned transport maps when the target distribution is concentrated. Plug-and-play (PnP) methods explicitly leverage the physical model to guide a learned prior, improving data consistency but reintroducing costly iterative computations. We instead propose AdjointFlow, a framework built on stochastic interpolants that injects the known physics into the generative transport itself. We design a data-dependent, adjoint-based initialization to reduce transport distance to the target, modify the transport velocity to remain regular as the endpoint distributions concentrate, and parameterize the network to build the initialization directly into the learned velocity. Together, these enable accurate reconstruction at substantially lower cost than generative models starting from Gaussian noise. Experiments on inverse scattering, seismic imaging, and black hole imaging show that AdjointFlow consistently produces more accurate reconstructions than conditional diffusion models while being up to five orders of magnitude faster at inference than PnP methods. Optional inference-time refinement that queries the physical model can further improve data consistency, matching or exceeding the accuracy of PnP methods without retraining.

## Equation Recast for Canonical Operator Learning Across Parametric PDEs

DevFeed: [Equation Recast for Canonical Operator Learning Across Parametric PDEs](<https://devfeed.tech/articles/equation-recast-for-canonical-operator-learning-across-parametric-pdes-128572.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2609.02982>)

Author: Qiyun Cheng, Valentin Duruisseaux, Cesar F. Clauser, Md Hossain Sahadath, Huihua Yang, Shaowu Pan, Nathaniel Ferraro, Anima Anandkumar, Wei Ji, Cristina Rea

Published: 2026-11-11T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Physics-guided deep learning](<https://devfeed.tech/topics/physics-guided-deep-learning.md>)

Tags: [canonical](<https://devfeed.tech/tags/canonical.md>), [convergence](<https://devfeed.tech/tags/convergence.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [data](<https://devfeed.tech/tags/data.md>), [data-driven](<https://devfeed.tech/tags/data-driven.md>), [data-efficiency](<https://devfeed.tech/tags/data-efficiency.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [fusion](<https://devfeed.tech/tags/fusion.md>), [neural](<https://devfeed.tech/tags/neural.md>), [operator-learning](<https://devfeed.tech/tags/operator-learning.md>), [physics-comp-ph](<https://devfeed.tech/tags/physics-comp-ph.md>), [physics-plasm-ph](<https://devfeed.tech/tags/physics-plasm-ph.md>), [zero-shot](<https://devfeed.tech/tags/zero-shot.md>)

### AI overview

Equation recast turns parametric PDE operator learning into learning one canonical operator: parameter-driven changes in the governing equation are derived analytically and folded into effective sources. The abstract reports zero-shot prediction in new parameter regimes, extrapolation across multi-parameter, nonlinear, and singular PDE settings, and integration of sparse heterogeneous datasets in a shared representation. Loss of convergence serves as a warning that the recast iteration may be failing. In high-fidelity tokamak simulations, canonical-domain mapping unifies electron-temperature data from four device geometries in one jointly trained operator. These are abstract-level claims; no quantitative comparisons or detailed implementation steps are provided.

### Source excerpt

arXiv:2609.02982v2 Announce Type: replace Abstract: Learning solution operators across broad parameter ranges can require substantial coverage of both input functions and physical parameters, particularly for purely data-driven parametric models. In addition, the resulting models may fail silently outside the training distribution. We introduce equation recast, which reformulates parametric operator learning as the learning of a single canonical operator. Parameter-induced operator variations are derived analytically from the governing equation and absorbed into effective sources, enabling zero-shot prediction across new parameter regimes. Across multi-parameter, nonlinear, and singular PDE settings, equation recast supports extrapolation, integrates sparse heterogeneous datasets in a shared canonical representation, and uses loss of convergence as an internal warning signal for failure of the recast iteration. In high-fidelity tokamak simulations for nuclear fusion, the framework unifies electron-temperature data across four device geometries through canonical-domain mapping within one jointly trained operator. Equation recast provides a route toward reusable neural PDE solvers combining equation-guided transfer, data efficiency, and monitorable inference.

## Deep Learning-Based Tracking and Lineage Reconstruction of Ligament Breakup

DevFeed: [Deep Learning-Based Tracking and Lineage Reconstruction of Ligament Breakup](<https://devfeed.tech/articles/deep-learning-based-tracking-and-lineage-reconstruction-of-ligament-breakup-123115.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2604.08711>)

Author: Vrushank Ahire, Vivek Kurumanghat, Mudasir Ganaie, Lipika Kabiraj

Published: 2026-11-09T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Deep learning](<https://devfeed.tech/topics/deep-learning.md>), [Image processing](<https://devfeed.tech/topics/image-processing.md>), [data augmentation](<https://devfeed.tech/topics/data-augmentation.md>), [Physics-guided deep learning](<https://devfeed.tech/topics/physics-guided-deep-learning.md>)

Tags: [analysis](<https://devfeed.tech/tags/analysis.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [deep-learning](<https://devfeed.tech/tags/deep-learning.md>), [droplets](<https://devfeed.tech/tags/droplets.md>), [fragmentation](<https://devfeed.tech/tags/fragmentation.md>), [framework](<https://devfeed.tech/tags/framework.md>), [one-to-many](<https://devfeed.tech/tags/one-to-many.md>), [physics](<https://devfeed.tech/tags/physics.md>), [robustness](<https://devfeed.tech/tags/robustness.md>), [size](<https://devfeed.tech/tags/size.md>), [statistics](<https://devfeed.tech/tags/statistics.md>), [synthetic-data-generation](<https://devfeed.tech/tags/synthetic-data-generation.md>), [tracking](<https://devfeed.tech/tags/tracking.md>), [transformer](<https://devfeed.tech/tags/transformer.md>)

### AI overview

The two-stage framework detects droplets and ligaments in high-speed images, then tracks their relationships while allowing one ligament to fragment into multiple objects. Its Faster R-CNN detector, trained with morphology-preserving synthetic data, reached a held-out F1 score of up to 0.872 across fourteen original-to-synthetic data configurations; robustness was tested across atomization conditions and liquid compositions. A Transformer-augmented multilayer perceptron classifies continuation, fragmentation, or non-association using physics-informed geometric features. On the held-out test set, it achieved 86.1% accuracy and 93.2% precision, with perfect recall for fragmentation despite severe class imbalance. The approach preserves parent-child lineage to support breakup statistics such as fragment multiplicity and droplet size distributions. Results described are for recordings of two impinging Carbopol gel jets.

### Source excerpt

arXiv:2604.08711v2 Announce Type: replace-cross Abstract: The disintegration of liquid sheets into ligaments and droplets involves highly transient, multi-scale dynamics that are complex to quantify from high-speed shadowgraphy images. Identification of droplets, ligaments, and blobs formed by the breakup of a liquid sheet, and their tracking from frame to frame, is an essential part of spray analysis. Analysis of such processes poses a fundamental challenge for conventional multi-object tracking frameworks, which enforce strict one-to-one temporal associations and cannot represent one-to-many fragmentation events. We present a two-stage deep learning framework that detects droplets and ligaments and identifies their temporal relationships across successive frames, modeling ligament deformation, fragmentation, and parent-child lineage. In the first stage, a Faster R-CNN detector with a ResNet-50 backbone and Feature Pyramid Network classifies ligaments and droplets in recordings of two impinging Carbopol gel jets. A morphology-preserving synthetic data generation strategy enriches the training set without physically implausible configurations, achieving a held-out test F1 score of up to 0.872 across fourteen original-to-synthetic data configurations, with robustness tested on different atomization conditions and liquid compositions. In the second stage, a Transformer-augmented multilayer perceptron classifies inter-frame object associations into continuation, fragmentation (one-to-many), and non-association using physics-informed geometric features. Despite severe class imbalance, the model achieves 86.1% accuracy, 93.2% precision, and perfect recall (1.00) for fragmentation events on the held-out test set. Together, the two stages preserve parent-child lineage and enable extraction of breakup statistics, including fragment multiplicity and droplet size distributions, offering an automated analysis of the primary mode of atomization.

## KolTa: A Physics-Informed Active Learning Framework for Formulation Discovery with Limited Data

DevFeed: [KolTa: A Physics-Informed Active Learning Framework for Formulation Discovery with Limited Data](<https://devfeed.tech/articles/kolta-a-physics-informed-active-learning-framework-for-formulation-discovery-with-limited-data-123250.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.06394>)

Author: Boyong Tang, Lei Yang, Tao Wang

Published: 2026-11-09T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [AI for science](<https://devfeed.tech/topics/ai-for-science.md>), [Physics-guided deep learning](<https://devfeed.tech/topics/physics-guided-deep-learning.md>)

Tags: [about](<https://devfeed.tech/tags/about.md>), [abstract](<https://devfeed.tech/tags/abstract.md>), [acquisition](<https://devfeed.tech/tags/acquisition.md>), [area](<https://devfeed.tech/tags/area.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [best](<https://devfeed.tech/tags/best.md>), [budget](<https://devfeed.tech/tags/budget.md>), [can](<https://devfeed.tech/tags/can.md>), [combines](<https://devfeed.tech/tags/combines.md>), [comparisons](<https://devfeed.tech/tags/comparisons.md>), [component](<https://devfeed.tech/tags/component.md>), [confidence](<https://devfeed.tech/tags/confidence.md>), [contributions](<https://devfeed.tech/tags/contributions.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [data](<https://devfeed.tech/tags/data.md>), [discovery](<https://devfeed.tech/tags/discovery.md>), [framework](<https://devfeed.tech/tags/framework.md>), [learning](<https://devfeed.tech/tags/learning.md>), [physics](<https://devfeed.tech/tags/physics.md>), [physics-informed](<https://devfeed.tech/tags/physics-informed.md>), [with](<https://devfeed.tech/tags/with.md>)

### AI overview

KolTa is an active learning framework for finding high-performing formulations under limited experimental budgets. It predicts component contributions, role-conditioned interactions, and temperature response, then selects batches with PRC-UCB, combining predicted utility, ensemble disagreement, and a target-temperature physics residual. In tests on three measured tasks and one simulated task, KolTa ranked first or second in mean area under the learning curve on every task. It beat seven of eight comparator strategies in paired tests, winning 75 of 78 comparisons against those seven. On the Drug task, an estimated median of 348 additional queries reached 80% average recovery of remaining targets across drugs, about 18% fewer than the fastest comparator. These results are limited to the four evaluated tasks.

### Source excerpt

arXiv:2611.06394v2 Announce Type: replace Abstract: Finding high-performing formulations is difficult when few have been measured and experimental budgets are limited. We propose KolTa, an active learning framework that selects batches to discover the top 5% of formulations at a target temperature under a fixed query budget. KolTa combines a three-stage predictor of component contributions, role-conditioned interactions, and anchored temperature response with task-specific soft physics constraints. Its Physics-Residual-Calibrated Upper Confidence Bound (PRC-UCB) acquisition rule combines predicted utility, ensemble disagreement, and a target-temperature physics residual, with nonnegative exploration weights fitted to out-of-fold signed errors. We evaluate the framework on three measured tasks and one simulated task covering binary-mixture viscosity, drug solubility, ionic-liquid viscosity, and electrolyte conductivity. KolTa ranks first or second in mean area under the learning curve (AULC) on every task and outperforms seven of the eight comparator strategies in paired tests, winning 75 of 78 paired comparisons across tasks and seeds against those seven strategies. Across the three measured tasks, its largest gap to the best mean AULC is 0.032, compared with at least 0.056 for every comparator. On Drug, reaching 80% average recovery of the remaining targets across drugs requires an estimated median of 348 additional formulation queries, about 18% fewer than the fastest comparator. Researchers can adapt the framework to diverse formulation problems through component roles and physical constraints.

## Stable probabilistic emulation of partially observed dynamical systems

DevFeed: [Stable probabilistic emulation of partially observed dynamical systems](<https://devfeed.tech/articles/stable-probabilistic-emulation-of-partially-observed-dynamical-systems-123394.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.09317>)

Author: Mario Massy, Sara Shamekh, Fabrizio Falasca, Matthieu Blanke

Published: 2026-11-09T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Climate / Weather / Ocean Modeling](<https://devfeed.tech/topics/climate-weather-ocean-modeling.md>), [Simulation](<https://devfeed.tech/topics/simulation.md>), [Physics-guided deep learning](<https://devfeed.tech/topics/physics-guided-deep-learning.md>)

Tags: [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [emulation](<https://devfeed.tech/tags/emulation.md>), [probabilistic](<https://devfeed.tech/tags/probabilistic.md>), [stable](<https://devfeed.tech/tags/stable.md>), [systems](<https://devfeed.tech/tags/systems.md>)

### AI overview

The paper introduces forced-dissipative stochastic interpolants for learning stochastic reduced dynamics in partially observed systems. The architecture imposes structural constraints inspired by turbulent dynamics to provide stability while retaining flexibility to model probabilistic behavior. The authors establish a Foster-Lyapunov drift condition, prove geometric ergodicity, and show stability under steady external forcing. On the two-timescale Lorenz-96 system, the emulator remains stable over long integrations and finite-amplitude forcing, and accurately predicts forced responses in both mean and variance. Evidence here is limited to the abstract.

### Source excerpt

arXiv:2611.09317v1 Announce Type: new Abstract: Predicting the long-term behavior of partially observed dynamical systems such as the coupled atmosphere-ocean system is central to societal decision-making. Motivated by the success of deep learning emulators in weather forecasting, data-driven climate models represent a promising approach. However, climate emulation poses distinct challenges: models must account for unresolved fast variables introduced by unavoidable coarse-graining, reproduce long-term statistics, and remain stable over arbitrarily long rollouts and under external perturbations. These requirements motivate stochastic data-driven emulators, but modern generative models lack stability guarantees. We introduce forced-dissipative stochastic interpolants, a generative framework for learning stochastic reduced dynamics. Its architecture uses structural constraints inspired by turbulent dynamics, which provide stability by design while retaining the flexibility of stochastic interpolants to represent complex probabilistic dynamics. Building on the theory of Markov-chain stability, we establish a Foster--Lyapunov drift condition and prove geometric ergodicity for our emulator, together with stability under steady external forcing. We evaluate the framework on the two-timescale Lorenz-96 system, a prototype of partially observed multiscale turbulence. The resulting emulator remains stable over long integrations and under finite-amplitude forcing, with accurate prediction of forced responses of both the mean and variance.

## Learning Generative Dynamics for 3D Molecule Generation via Sequential Neuro-Symbolic Constraints

DevFeed: [Learning Generative Dynamics for 3D Molecule Generation via Sequential Neuro-Symbolic Constraints](<https://devfeed.tech/articles/learning-generative-dynamics-for-3d-molecule-generation-via-sequential-neuro-symbolic-constraints-123362.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.09142>)

Author: Zhuoran Li, Xu Sun, Chang Wen Chen, Wanyu Lin

Published: 2026-11-09T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Physics-guided deep learning](<https://devfeed.tech/topics/physics-guided-deep-learning.md>), [Model Development](<https://devfeed.tech/topics/model-development.md>)

Tags: [3d](<https://devfeed.tech/tags/3d.md>), [adaptive](<https://devfeed.tech/tags/adaptive.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [denoising](<https://devfeed.tech/tags/denoising.md>), [diffusion](<https://devfeed.tech/tags/diffusion.md>), [diffusion-models](<https://devfeed.tech/tags/diffusion-models.md>), [domain](<https://devfeed.tech/tags/domain.md>), [generation](<https://devfeed.tech/tags/generation.md>), [generative](<https://devfeed.tech/tags/generative.md>), [learning](<https://devfeed.tech/tags/learning.md>), [modeling](<https://devfeed.tech/tags/modeling.md>), [models](<https://devfeed.tech/tags/models.md>), [neural](<https://devfeed.tech/tags/neural.md>), [principles](<https://devfeed.tech/tags/principles.md>), [real-world](<https://devfeed.tech/tags/real-world.md>), [rules](<https://devfeed.tech/tags/rules.md>), [tasks](<https://devfeed.tech/tags/tasks.md>), [training](<https://devfeed.tech/tags/training.md>), [user](<https://devfeed.tech/tags/user.md>)

### AI overview

SensDiff incorporates symbolic physicochemical constraints into 3D molecule diffusion training, addressing the risk that data-learned atomic distributions produce chemically invalid samples. It follows a coarse-to-fine trajectory: constraints with adaptive weights first address large-scale geometric conflicts, then guide iterative denoising toward chemical consistency. The framework also presents domain knowledge as interpretable controls. The abstract reports improved chemical validity while adhering to physicochemical rules and user-specified targets on molecular benchmarks and constrained-generation tasks; it provides no quantitative results or comparison details.

### Source excerpt

arXiv:2611.09142v1 Announce Type: new Abstract: Diffusion models have recently emerged as a powerful paradigm for generating 3D molecules. Despite showing remarkable promise, they learn atomic distributions implicitly from data, without explicitly encoding physicochemical rules into their generative dynamics. Consequently, generated samples can be statistically plausible yet chemically invalid. To address this problem, we propose Sequential Neuro-Symbolic Constrained Diffusion (SensDiff), a framework that integrates symbolic rules into generative modeling, enabling the model to internalize scientific priors during training. Central to SensDiff is the observation that the diffusion trajectory of 3D molecule generation exhibits a coarse-to-fine evolution: valid molecules arise by first mitigating macroscopic geometric conflicts and then refining structures toward microscopic chemical consistency through iterative denoising. Accordingly, SensDiff mirrors this dynamic by sequentially enforcing constraints with adaptive weighting during training, progressively steering generation toward valid molecular geometries. Moreover, SensDiff translates domain knowledge into interpretable generative controls, grounding opaque neural denoising in scientific principles. Experiments on molecular benchmarks and real-world constrained generation tasks confirm that SensDiff consistently improves chemical validity while adhering to physicochemical rules and user-specified targets. The code is publicly available at https://github.com/WanyuGroup/NeurIPS2026_SensDiff.

## Equivariant Deep Learning in Direct-Product Space for Electronic-Structure Hamiltonian Prediction

DevFeed: [Equivariant Deep Learning in Direct-Product Space for Electronic-Structure Hamiltonian Prediction](<https://devfeed.tech/articles/equivariant-deep-learning-in-direct-product-space-for-electronic-structure-hamiltonian-prediction-120234.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.07695>)

Author: Shi Yin, Zujian Dai, Xinyang Pan, Lixin He

Published: 2026-11-06T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Physics-guided deep learning](<https://devfeed.tech/topics/physics-guided-deep-learning.md>), [Representation learning](<https://devfeed.tech/topics/representation-learning.md>)

Tags: [architecture](<https://devfeed.tech/tags/architecture.md>), [attention](<https://devfeed.tech/tags/attention.md>), [cond-mat-mtrl-sci](<https://devfeed.tech/tags/cond-mat-mtrl-sci.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [dataset](<https://devfeed.tech/tags/dataset.md>), [deep-learning](<https://devfeed.tech/tags/deep-learning.md>), [efficiency](<https://devfeed.tech/tags/efficiency.md>), [hamiltonian](<https://devfeed.tech/tags/hamiltonian.md>), [invariant](<https://devfeed.tech/tags/invariant.md>), [learning](<https://devfeed.tech/tags/learning.md>), [message](<https://devfeed.tech/tags/message.md>), [physics-comp-ph](<https://devfeed.tech/tags/physics-comp-ph.md>), [prediction](<https://devfeed.tech/tags/prediction.md>), [spatial](<https://devfeed.tech/tags/spatial.md>)

### AI overview

DP-Learner predicts electronic-structure Hamiltonians by passing messages directly between Hamiltonian-like operator blocks in the direct-product orbital and spin representation, avoiding the explicit Clebsch-Gordan decomposition and reconstruction used by the representation strategy described for existing architectures. Its proposed Lie-generator directional embeddings and invariantly weighted products and contractions preserve equivariance to SO(3), SU(2), and spatial inversions. A Transformer uses invariant scalar routing to gate matrix units, directional propagation, attention, and environment context interactions. The authors also introduce MP-HAM-110K, with over 110,000 material structures across 68 elements and Hamiltonian and wavefunction data. They report improved accuracy and inference efficiency on MP-HAM-110K and Materials-HAM-SOC; the abstract gives no quantitative results.

### Source excerpt

arXiv:2611.07695v1 Announce Type: cross Abstract: Most existing equivariant architectures for Hamiltonian prediction propagate features in direct-sum spaces of irreducible spherical tensors, whereas the target Hamiltonian is naturally an operator block in a direct-product representation over local orbital and spin bases. This representation mismatch fragments contiguous Hamiltonian blocks, obscures direct manipulation of orbital couplings, and organizes features into irregular angular-momentum components that require numerous transformations and are poorly aligned with efficient hardware execution and memory management. To address both the representation mismatch and computational bottleneck, we propose DP-Learner, an equivariant learning framework that performs message passing directly on Hamiltonian-like operator blocks in the direct-product space. Theoretically, we propose Lie-generator-based directional embedding and invariantly weighted left--right products and contractions that preserve equivariance to SO(3), SU(2), and spatial inversions, without explicit Clebsch--Gordan decomposition or inverse reconstruction. Methodologically, we instantiate this principle as a Transformer architecture, in which Hamiltonian-like feature matrices or tensors evolve through equivariant operators in the direct-product space, while non-linear invariant scalar routing controls structured gating over matrix units, directional propagation, attention, and environment-embedded context interactions. We further construct MP-HAM-110K, a large-scale Hamiltonian dataset comprising more than 110,000 material structures covering 68 elements with both Hamiltonian and wavefunction data. Experiments on MP-HAM-110K and Materials-HAM-SOC demonstrate that DP-Learner dramatically improves Hamiltonian prediction accuracy and inference efficiency over the state-of-the-art method.

[Next page](<https://devfeed.tech/topics/physics-guided-deep-learning.md?cursor=WyIyMDI2LTExLTA2VDA1OjAwOjAwKzAwOjAwIiwgImRhOWYzN2QwLTM1ZDQtNDg5OS1iYWQ0LWI5YjE5YTNkZWE3OSJd>)