Metraipolis › archive › dc_04463e26d8cc
Learning Chance-Constrained MDPs with Bellman Distributional Certificates — arXiv:2609.30856
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.30856
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T02:10:58.211Z
- SHA-256 of the archived text
209a2bc60a3922c38bea2b4672b44442b8342262471c54b13d9a723cada2c42c- Stable URL of this page
https://metraipolis.com/documento/dc_04463e26d8cc
How to cite
Learning Chance-Constrained MDPs with Bellman Distributional Certificates — arXiv:2609.30856 — arXiv (en), https://arxiv.org/abs/2609.30856 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_04463e26d8cc (Metraipolis archive dc_04463e26d8cc, sha256 209a2bc60a3922c38bea2b4672b44442b8342262471c54b13d9a723cada2c42c)
BibTeX
@misc{dc04463e26d8cc,
title = {Learning Chance-Constrained MDPs with Bellman Distributional Certificates — arXiv:2609.30856},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.30856},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_04463e26d8cc; sha256 209a2bc60a3922c38bea2b4672b44442b8342262471c54b13d9a723cada2c42c},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.30856. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Safe reinforcement learning (RL) commonly enforces expected-cost constraints, but such expectation safety may fail to control the probability of rare high-cost trajectories. Chance-constrained MDPs (CCMDPs) impose a stronger probability-level requirement, but are widely viewed as harder because the chance constraint is nonconvex and depends on the full trajectory rather than a Bellman-linear expectation. In this paper, we reveal that this computational difficulty does not necessarily imply a higher statistical price. For tabular discounted CCMDPs with fixed bounded successor support and access to a certified planning oracle, we establish a model-based upper bound, with a matching lower bound up to logarithmic terms. Technically, our key idea is the \emph{Bellman distributional certificate}, which constructs a Bellman recursion for constraint violation probabilities before policy selection. The certificate can be reused across candidate policies; combined with shared row-wise reverse-KL confidence sets, it gives a policy-uniform trajectory-KL transfer without a union bound over policies or time--budget Bellman tables. For stochastic policies, we give a model-free variance-reduced policy-gradient algorithm with a finite-sample expected KKT-residual guarantee and independent validation of every accepted policy. Numerical experiments on synthetic CCMDPs and an IEEE 14-bus energy storage control benchmark illustrate the safety and mechanism behavior of the proposed algorithms.
Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-28 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.30856 — 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 2026-10-08T01:43:06.709Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).