Near-Minimax-Optimal Distributional Reinforcement Learning with a Generative Model
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866916465289986048 |
|---|---|
| author | Rowland, Mark Wenliang, Li Kevin Munos, Rémi Lyle, Clare Tang, Yunhao Dabney, Will |
| author_facet | Rowland, Mark Wenliang, Li Kevin Munos, Rémi Lyle, Clare Tang, Yunhao Dabney, Will |
| contents | We propose a new algorithm for model-based distributional reinforcement learning (RL), and prove that it is minimax-optimal for approximating return distributions with a generative model (up to logarithmic factors), resolving an open question of Zhang et al. (2023). Our analysis provides new theoretical results on categorical approaches to distributional RL, and also introduces a new distributional Bellman equation, the stochastic categorical CDF Bellman equation, which we expect to be of independent interest. We also provide an experimental study comparing several model-based distributional RL algorithms, with several takeaways for practitioners. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_07598 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Near-Minimax-Optimal Distributional Reinforcement Learning with a Generative Model Rowland, Mark Wenliang, Li Kevin Munos, Rémi Lyle, Clare Tang, Yunhao Dabney, Will Machine Learning We propose a new algorithm for model-based distributional reinforcement learning (RL), and prove that it is minimax-optimal for approximating return distributions with a generative model (up to logarithmic factors), resolving an open question of Zhang et al. (2023). Our analysis provides new theoretical results on categorical approaches to distributional RL, and also introduces a new distributional Bellman equation, the stochastic categorical CDF Bellman equation, which we expect to be of independent interest. We also provide an experimental study comparing several model-based distributional RL algorithms, with several takeaways for practitioners. |
| title | Near-Minimax-Optimal Distributional Reinforcement Learning with a Generative Model |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2402.07598 |