Brain-Like Replay Naturally Emerges in Reinforcement Learning Agents
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_ | 1866909337915490304 |
|---|---|
| author | Wang, Jiyi Tang, Likai Chen, Huimiao Mattar, Marcelo G Song, Sen |
| author_facet | Wang, Jiyi Tang, Likai Chen, Huimiao Mattar, Marcelo G Song, Sen |
| contents | Replay is a powerful strategy to promote learning in artificial intelligence and the brain. However, the conditions to generate it and its functional advantages have not been fully recognized. In this study, we develop a modular reinforcement learning model that could generate replay. We prove that replay generated in this way helps complete the task. We also analyze the information contained in the representation and provide a mechanism for how replay makes a difference. Our design avoids complex assumptions and enables replay to emerge naturally within a task-optimized paradigm. Our model also reproduces key phenomena observed in biological agents. This research explores the structural biases in modular ANN to generate replay and its potential utility in developing efficient RL. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_01467 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Brain-Like Replay Naturally Emerges in Reinforcement Learning Agents Wang, Jiyi Tang, Likai Chen, Huimiao Mattar, Marcelo G Song, Sen Systems and Control Artificial Intelligence Computational Engineering, Finance, and Science Neural and Evolutionary Computing Neurons and Cognition Replay is a powerful strategy to promote learning in artificial intelligence and the brain. However, the conditions to generate it and its functional advantages have not been fully recognized. In this study, we develop a modular reinforcement learning model that could generate replay. We prove that replay generated in this way helps complete the task. We also analyze the information contained in the representation and provide a mechanism for how replay makes a difference. Our design avoids complex assumptions and enables replay to emerge naturally within a task-optimized paradigm. Our model also reproduces key phenomena observed in biological agents. This research explores the structural biases in modular ANN to generate replay and its potential utility in developing efficient RL. |
| title | Brain-Like Replay Naturally Emerges in Reinforcement Learning Agents |
| topic | Systems and Control Artificial Intelligence Computational Engineering, Finance, and Science Neural and Evolutionary Computing Neurons and Cognition |
| url | https://arxiv.org/abs/2402.01467 |