Metraipolis › archive › dc_41609ce834da
Last-Iterate Guarantees for Online Reinforcement Learning in Structured Constrained MDPs — arXiv:2609.32933
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.32933
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T22:36:59.646Z
- SHA-256 of the archived text
fe48c2ffd5c3e9cd920a029b0b626b995708a83877c33d322d8c32800e7313a4- Stable URL of this page
https://metraipolis.com/documento/dc_41609ce834da
How to cite
Last-Iterate Guarantees for Online Reinforcement Learning in Structured Constrained MDPs — arXiv:2609.32933 — arXiv (en), https://arxiv.org/abs/2609.32933 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_41609ce834da (Metraipolis archive dc_41609ce834da, sha256 fe48c2ffd5c3e9cd920a029b0b626b995708a83877c33d322d8c32800e7313a4)
BibTeX
@misc{dc41609ce834da,
title = {Last-Iterate Guarantees for Online Reinforcement Learning in Structured Constrained MDPs — arXiv:2609.32933},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.32933},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_41609ce834da; sha256 fe48c2ffd5c3e9cd920a029b0b626b995708a83877c33d322d8c32800e7313a4},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.32933. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
In safety-critical applications, deployment uses a single policy, whose performance and constraint satisfaction should hold directly rather than only for an average or mixture of training policies. This motivates last-iterate guarantees in constrained reinforcement learning. Recent progress has established such guarantees in exact-gradient or tabular online settings, yet scalable results for structured large-state problems remain open. We develop a general, statistically efficient framework for last-iterate convergence in structured Constrained MDPs (CMDPs). Our analysis separates contraction of the regularised primal-dual dynamics from actor approximation and statistical errors in policy evaluation under online exploration. This enables model-free on- and off-policy learning with structured function approximation: optimistic policy evaluation avoids explicit transition-model construction, while a compact parametric actor avoids maintaining mixtures or histories of past policies. We instantiate the framework for linear CMDPs and general function approximation, obtaining representation-dependent complexity and improved target-accuracy dependence over prior optimistic regularised primal-dual analyses. We further validate the stabilising effect predicted by our theory on a synthetic linear CMDP: the regularised method exhibits stable last-iterate behaviour, whereas its unregularised counterpart shows larger oscillations. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.LG). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.32933 — 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 2026-10-07T22:14:27.813Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).