Metraipolis › archive › dc_744b228109ff
Future-Informed Actor-Critic — arXiv:2611.09417
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.09417
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-09T22:41:35.857Z
- SHA-256 of the archived text
53be8a8d3eea8e2cc6fbcc4c692a8bb174d9c8e377d5c94e3db98153e525b808- Stable URL of this page
https://metraipolis.com/documento/dc_744b228109ff
How to cite
Future-Informed Actor-Critic — arXiv:2611.09417 — arXiv (en), https://arxiv.org/abs/2611.09417 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_744b228109ff (Metraipolis archive dc_744b228109ff, sha256 53be8a8d3eea8e2cc6fbcc4c692a8bb174d9c8e377d5c94e3db98153e525b808)
BibTeX
@misc{dc744b228109ff,
title = {Future-Informed Actor-Critic — arXiv:2611.09417},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.09417},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_744b228109ff; sha256 53be8a8d3eea8e2cc6fbcc4c692a8bb174d9c8e377d5c94e3db98153e525b808},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.09417. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Actor-critic methods optimize a parametrized policy using a critic to estimate their policy gradient. Recent work has shown that conditioning the critic on features of future states, actions, or rewards can improve sample efficiency and even final return. However, existing methods are restricted to exogenous features (independent of actions), overlooking endogenous features (dependent on actions) that may be crucial for credit assignment. We introduce the future-informed value function, which can be conditioned on both exogenous and endogenous features of future states, actions, and rewards. Then, we express the policy gradient using this value function, yielding the future-informed policy gradient. Building on that, we propose the future-informed actor-critic algorithm. Empirical results show that this algorithm leveraging endogenous features of the future outperforms standard actor-critic algorithms in environments with noisy and sparse rewards. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2611.09417 — 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 2026-11-09T22:33:47.159Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).