# Simulated petrological labels enable probabilistic mineralogy prediction from whole-rock geochemistry

DevFeed: [Simulated petrological labels enable probabilistic mineralogy prediction from whole-rock geochemistry](<https://devfeed.tech/articles/simulated-petrological-labels-enable-probabilistic-mineralogy-prediction-from-whole-rock-geochemistry-103954.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.36429>)

Author: Matthew J. Liu, Aaron D. Kaplan, Michael L. Whittaker, Kristin A. Persson

Published: 2026-10-28T04:00:00Z

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>), [Self-supervised learning](<https://devfeed.tech/topics/self-supervised-learning.md>), [Inference](<https://devfeed.tech/topics/inference.md>), [few-shot](<https://devfeed.tech/topics/few-shot.md>)

Tags: [000](<https://devfeed.tech/tags/000.md>), [abstract](<https://devfeed.tech/tags/abstract.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [can](<https://devfeed.tech/tags/can.md>), [categories](<https://devfeed.tech/tags/categories.md>), [challenges](<https://devfeed.tech/tags/challenges.md>), [component](<https://devfeed.tech/tags/component.md>), [composition](<https://devfeed.tech/tags/composition.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [data](<https://devfeed.tech/tags/data.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [distributions](<https://devfeed.tech/tags/distributions.md>), [error](<https://devfeed.tech/tags/error.md>), [express](<https://devfeed.tech/tags/express.md>), [full](<https://devfeed.tech/tags/full.md>), [guides](<https://devfeed.tech/tags/guides.md>), [improvement](<https://devfeed.tech/tags/improvement.md>), [industrial](<https://devfeed.tech/tags/industrial.md>), [layer](<https://devfeed.tech/tags/layer.md>), [prediction](<https://devfeed.tech/tags/prediction.md>), [probabilistic](<https://devfeed.tech/tags/probabilistic.md>)

## AI overview

A model trained with simulated petrological labels predicts probabilistic mineral assemblages from whole-rock chemistry, addressing the scarcity of measured labels and the fact that one composition can reflect multiple formation conditions. It combines a transformer encoder self-supervised on roughly 788,000 compositions, prediction heads trained on nearly one million simulated labels, and a recalibration layer fitted to measured mineralogy. On 640 volcanic rocks, zero-shot predictions distinguished the presence of most mineral categories and ranked relative richness, but predicted abundances were systematically offset. Recalibration with 25 labeled rocks captured 76% of the likelihood improvement achieved with 368 fitting rocks; abundance error improved more slowly, from 11.0% zero-shot to 9.5% with the full fitting set. The results show both the value and bias limitations of simulator-supervised prediction.

## Source excerpt

arXiv:2610.36429v1 Announce Type: new Abstract: Modal mineralogy, the mineral identities and abundances in rocks, guides geological studies and industrial resource extraction. While measuring modal mineralogy is time-intensive, whole-rock chemical measurements furnish elemental concentrations at higher throughput. Predicting mineralogy from elemental data, however, poses a one-to-many inverse problem, and labeled mineralogical datasets are too sparse to train on at scale. We address both challenges by simulating petrological labels that express multiple formation conditions per composition and at a scale unattainable by measurements. Nearly one million simulated labels supervise the middle component of a three-part model: a transformer encoder self-supervised on roughly 788,000 whole-rock compositions, the simulation-trained heads that predict distributions over mineral assemblages, and a recalibration layer that aligns the predictions with measured mineralogy. Zero-shot on 640 volcanic rocks, the model discriminates the presence of most mineral categories and predicts which rocks are richer in each mineral. However, the predicted abundances are systematically offset from measured values. Recalibration fit on as few as 25 labeled rocks recovers 76 percent of the likelihood improvement from fitting on the full 368-rock fitting set, while abundance error responds more slowly (11.0% zero-shot and 9.5% with the full fitting set). This work demonstrates how a machine learning model in the geochemical sciences can both leverage an imperfect simulator and mitigate its systematic biases.