Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | https://arxiv.org/abs/2509.26000 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866910012552511488 |
|---|---|
| author | Ebi, Daniel Lambrechts, Gaspard Ernst, Damien Böhm, Klemens |
| author_facet | Ebi, Daniel Lambrechts, Gaspard Ernst, Damien Böhm, Klemens |
| contents | Asymmetric actor-critic methods are widely used in partially observable reinforcement learning, but typically assume full state observability to condition the critic during training, which is often unrealistic in practice. We introduce the informed asymmetric actor-critic framework, allowing the critic to be conditioned on arbitrary state-dependent privileged signals without requiring access to the full state. We show that any such privileged signal yields unbiased policy gradient estimates, substantially expanding the set of admissible privileged information. This raises the problem of selecting the most adequate privileged information in order to improve learning. For this purpose, we propose two novel informativeness criteria: a dependence-based test that can be applied prior to training, and a criterion based on improvements in value prediction accuracy that can be applied post-hoc. Empirical results on partially observable benchmark tasks and synthetic environments demonstrate that carefully selected privileged signals can match or outperform full-state asymmetric baselines while relying on strictly less state information. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_26000 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Informed Asymmetric Actor-Critic: Leveraging Privileged Signals Beyond Full-State Access Ebi, Daniel Lambrechts, Gaspard Ernst, Damien Böhm, Klemens Machine Learning Asymmetric actor-critic methods are widely used in partially observable reinforcement learning, but typically assume full state observability to condition the critic during training, which is often unrealistic in practice. We introduce the informed asymmetric actor-critic framework, allowing the critic to be conditioned on arbitrary state-dependent privileged signals without requiring access to the full state. We show that any such privileged signal yields unbiased policy gradient estimates, substantially expanding the set of admissible privileged information. This raises the problem of selecting the most adequate privileged information in order to improve learning. For this purpose, we propose two novel informativeness criteria: a dependence-based test that can be applied prior to training, and a criterion based on improvements in value prediction accuracy that can be applied post-hoc. Empirical results on partially observable benchmark tasks and synthetic environments demonstrate that carefully selected privileged signals can match or outperform full-state asymmetric baselines while relying on strictly less state information. |
| title | Informed Asymmetric Actor-Critic: Leveraging Privileged Signals Beyond Full-State Access |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2509.26000 |