# Probabilistic Signature Inversion: Learning Conditional Distributions from Truncated Signatures

DevFeed: [Probabilistic Signature Inversion: Learning Conditional Distributions from Truncated Signatures](<https://devfeed.tech/articles/probabilistic-signature-inversion-learning-conditional-distributions-from-truncated-signatures-132808.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2606.15332>)

Author: Junoh Kang, Kiseop Lee, Bohyung Han

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

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>), [Training AI Models](<https://devfeed.tech/topics/training-ai-models.md>), [Neural Network](<https://devfeed.tech/topics/neural-network.md>)

Tags: [analysis](<https://devfeed.tech/tags/analysis.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [code](<https://devfeed.tech/tags/code.md>), [conditional](<https://devfeed.tech/tags/conditional.md>), [continuous](<https://devfeed.tech/tags/continuous.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [data](<https://devfeed.tech/tags/data.md>), [distribution](<https://devfeed.tech/tags/distribution.md>), [distributions](<https://devfeed.tech/tags/distributions.md>), [error](<https://devfeed.tech/tags/error.md>), [estimator](<https://devfeed.tech/tags/estimator.md>), [experimental](<https://devfeed.tech/tags/experimental.md>), [feature](<https://devfeed.tech/tags/feature.md>), [financial](<https://devfeed.tech/tags/financial.md>), [financial-data](<https://devfeed.tech/tags/financial-data.md>), [flow](<https://devfeed.tech/tags/flow.md>), [flow-matching](<https://devfeed.tech/tags/flow-matching.md>), [form](<https://devfeed.tech/tags/form.md>), [framework](<https://devfeed.tech/tags/framework.md>), [learning](<https://devfeed.tech/tags/learning.md>), [probabilistic](<https://devfeed.tech/tags/probabilistic.md>)

## AI overview

Truncated-signature inversion is ill-posed because multiple paths can share a truncated signature, so the paper treats it as learning the conditional path distribution. It uses a signature-conditioned flow matching model and frames posterior-sampling reconstruction error as irreducible uncertainty after conditioning. For Gaussian process families, it derives this error under linear-statistics conditioning: a closed form for log-GBM and numerically tractable formulas for drifted fBM and OU. This provides an oracle upper bound on error under truncated-signature conditioning, which is more informative than linear statistics. Experiments report that the learned sampler closely matches the theoretical linear-statistics posterior, errors fall with richer truncated signatures, and generated paths preserve the conditioning signature and key temporal and distributional structure. Applicability is also demonstrated on financial data beyond the theoretically analyzed process families.

## Source excerpt

arXiv:2606.15332v2 Announce Type: replace Abstract: The signature transform is a principled feature map for continuous-time paths, valued for its uniqueness and universality. Recovering a path from its truncated signature is, however, structurally ill-posed, because the truncated signature map is not injective. We therefore reframe truncated signature inversion as a probabilistic problem---learning the conditional distribution of a path given its truncated signature---and adopt a signature-conditioned flow matching model as a practical estimator. This formulation elucidates the fundamental difficulty of inversion: posterior-sampling reconstruction error quantifies the irreducible uncertainty remaining after conditioning on a statistic. For Gaussian process families, we derive this error from the tractable posterior under linear-statistics conditioning, obtaining a closed form for log-GBM and numerically tractable formulas for drifted fBM and OU. This oracle error upper-bounds the corresponding error under truncated-signature conditioning, since truncated signatures are more informative than linear statistics. The experimental results are consistent with the theoretical analysis: the learned sampler under linear-statistics conditioning closely matches the theoretically derived posterior in both reconstruction error and distribution, while reconstruction errors decrease under richer truncated-signature conditioning. Moreover, generated paths faithfully recover the conditioning signature and preserve key distributional and temporal structures. Together, these results establish a well-posed probabilistic framework for truncated-signature inversion, with applicability demonstrated on financial data beyond the process families covered by theory. Code is available at https://github.com/Junoh-Kang/ProbSigInv.