Metraipolis › archive › dc_e1e8894370bc
An analysis of Mirror-Descent Soft Actor-Critic — arXiv:2609.35466
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.35466
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T09:33:30.012Z
- SHA-256 of the archived text
0dd679540aa159897c433042a47b588b8064286d67a525677a1ed133837e8810- Stable URL of this page
https://metraipolis.com/documento/dc_e1e8894370bc
How to cite
An analysis of Mirror-Descent Soft Actor-Critic — arXiv:2609.35466 — arXiv (en), https://arxiv.org/abs/2609.35466 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e1e8894370bc (Metraipolis archive dc_e1e8894370bc, sha256 0dd679540aa159897c433042a47b588b8064286d67a525677a1ed133837e8810)
BibTeX
@misc{dce1e8894370bc,
title = {An analysis of Mirror-Descent Soft Actor-Critic — arXiv:2609.35466},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.35466},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e1e8894370bc; sha256 0dd679540aa159897c433042a47b588b8064286d67a525677a1ed133837e8810},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.35466. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Soft Actor-Critic (SAC) is widely used for entropy-regularised reinforcement learning with continuous action spaces, and practical implementations perform only a few actor steps towards an evolving target. In this work, we prove convergence guarantees when the target policy arises from policy mirror descent and compare it with the classical Gibbs target. We derive sufficient conditions for the strong convexity and smoothness of the actor objective, characterised by the curvature of the $Q$-function estimate through the Legendre differential operator, and establish an $\mathcal{O}\!\left(N^{-\frac{1}{5}}\right)$ best-iterate finite-time convergence rate up to actor and critic approximation errors. Moreover, the mirror-descent step size $\lambda$ directly controls the target drift and hence actor tracking error, whereas the analogous Gibbs bound contains a non-vanishing tracking term.
Tema degli esploratori: statistica e scienza dei dati (elenco arXiv stat.ML, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Optimization and Control (math.OC); Machine Learning (stat.ML).
Pagina del lavoro: https://arxiv.org/abs/2609.35466 — 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 2026-10-06T08:24:13.973Z dalla pagina https://arxiv.org/list/stat.ML/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).