Metraipolis › archive › dc_85ee76e37261
Propensity Straight-Through Gradients for Discrete Stochastic Systems — arXiv:2608.25631
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.25631
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T08:39:38.042Z
- SHA-256 of the archived text
2965ea295c302ccd414475fb15715b7a7630e2ca51c8914f7c3e30185e19cd6c- Stable URL of this page
https://metraipolis.com/documento/dc_85ee76e37261
How to cite
Propensity Straight-Through Gradients for Discrete Stochastic Systems — arXiv:2608.25631 — arXiv (en), https://arxiv.org/abs/2608.25631 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_85ee76e37261 (Metraipolis archive dc_85ee76e37261, sha256 2965ea295c302ccd414475fb15715b7a7630e2ca51c8914f7c3e30185e19cd6c)
BibTeX
@misc{dc85ee76e37261,
title = {Propensity Straight-Through Gradients for Discrete Stochastic Systems — arXiv:2608.25631},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.25631},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_85ee76e37261; sha256 2965ea295c302ccd414475fb15715b7a7630e2ca51c8914f7c3e30185e19cd6c},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.25631. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Continuous-time Markov chains (CTMCs) provide the backbone for modeling discrete stochastic dynamics across applied, physical, and biological sciences. Their integration with modern gradient-based machine learning, however, is limited by the hard categorical event selection intrinsic to Gillespie-type simulation algorithms. We exploit the affine state update to obtain the exact one-step conditional-mean sensitivity by differentiating normalized reaction propensities. We pair this backward rule with exact forward trajectories to define the propensity straight-through (PST) estimator. At the trajectory level, we show that one-step sensitivities composed across events can depart from the exact multistep sensitivity. We derive the resulting per-step discrepancy in closed form and prove that it vanishes identically for affine downstream dependence. PST matches the accuracy of Gumbel-Softmax straight-through across all benchmarks: reversible dimerization (0.06% error), a genetic oscillator (1.7% error), a 50-task repressilator suite (0.17% median error), and patch-clamp ion-channel recordings ($R^2$ = 0.988). Under matched settings, PST converges 3.0-fold faster on the oscillator and 2.1-fold faster on the ion channel. At deep-learning scale, PST trains a 203,796-parameter stochastic reaction network with hard sampling, reaching 98.22% MNIST digit classification accuracy. By differentiating an exact conditional mean rather than a relaxed sample, PST offers a temperature- and Gumbel-free path to scalable gradient-based learning through exact stochastic trajectories. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-01 in cs.LG). Categorie arXiv: Quantitative Methods (q-bio.QM) ; Disordered Systems and Neural Networks (cond-mat.dis-nn); Machine Learning (cs.LG); Computational Physics (physics.comp-ph); Molecular Networks (q-bio.MN). Pagina del lavoro: https://arxiv.org/abs/2608.25631 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-17T08:13:26.873Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).