Generating individual intrinsic reward for cooperative multiagent reinforcement learning
Multiagent reinforcement learning holds considerable promise to deal with cooperative multiagent tasks. Unfortunately, the only global reward shared by all agents in the cooperative tasks may lead to the lazy agent problem. To cope with such a problem, we propose a generating individual intrinsic re...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SAGE Publishing
2021-10-01
|
| Series: | International Journal of Advanced Robotic Systems |
| Online Access: | https://doi.org/10.1177/17298814211044946 |
| _version_ | 1818395195134181376 |
|---|---|
| author | Haolin Wu Hui Li Jianwei Zhang Zhuang Wang Jianeng Zhang |
| author_facet | Haolin Wu Hui Li Jianwei Zhang Zhuang Wang Jianeng Zhang |
| author_sort | Haolin Wu |
| collection | DOAJ |
| description | Multiagent reinforcement learning holds considerable promise to deal with cooperative multiagent tasks. Unfortunately, the only global reward shared by all agents in the cooperative tasks may lead to the lazy agent problem. To cope with such a problem, we propose a generating individual intrinsic reward algorithm, which introduces an intrinsic reward encoder to generate an individual intrinsic reward for each agent and utilizes the hypernetworks as the decoder to help to estimate the individual action values of the decomposition methods based on the generated individual intrinsic reward. Experimental results in the StarCraft II micromanagement benchmark prove that the proposed algorithm can increase learning efficiency and improve policy performance. |
| first_indexed | 2024-12-14T06:13:14Z |
| format | Article |
| id | doaj.art-77deac9f7fcb4feb8bc5992f0579ef83 |
| institution | Directory Open Access Journal |
| issn | 1729-8814 |
| language | English |
| last_indexed | 2024-12-14T06:13:14Z |
| publishDate | 2021-10-01 |
| publisher | SAGE Publishing |
| record_format | Article |
| series | International Journal of Advanced Robotic Systems |
| spelling | doaj.art-77deac9f7fcb4feb8bc5992f0579ef832022-12-21T23:14:05ZengSAGE PublishingInternational Journal of Advanced Robotic Systems1729-88142021-10-011810.1177/17298814211044946Generating individual intrinsic reward for cooperative multiagent reinforcement learningHaolin Wu0Hui Li1Jianwei Zhang2Zhuang Wang3Jianeng Zhang4 College of Computer Science, Sichuan University, Chengdu, China National Key Laboratory of Fundamental Science on Synthetic Vision, Sichuan University, Chengdu, China National Key Laboratory of Fundamental Science on Synthetic Vision, Sichuan University, Chengdu, China College of Computer Science, Sichuan University, Chengdu, China College of Computer Science, Sichuan University, Chengdu, ChinaMultiagent reinforcement learning holds considerable promise to deal with cooperative multiagent tasks. Unfortunately, the only global reward shared by all agents in the cooperative tasks may lead to the lazy agent problem. To cope with such a problem, we propose a generating individual intrinsic reward algorithm, which introduces an intrinsic reward encoder to generate an individual intrinsic reward for each agent and utilizes the hypernetworks as the decoder to help to estimate the individual action values of the decomposition methods based on the generated individual intrinsic reward. Experimental results in the StarCraft II micromanagement benchmark prove that the proposed algorithm can increase learning efficiency and improve policy performance.https://doi.org/10.1177/17298814211044946 |
| spellingShingle | Haolin Wu Hui Li Jianwei Zhang Zhuang Wang Jianeng Zhang Generating individual intrinsic reward for cooperative multiagent reinforcement learning International Journal of Advanced Robotic Systems |
| title | Generating individual intrinsic reward for cooperative multiagent reinforcement learning |
| title_full | Generating individual intrinsic reward for cooperative multiagent reinforcement learning |
| title_fullStr | Generating individual intrinsic reward for cooperative multiagent reinforcement learning |
| title_full_unstemmed | Generating individual intrinsic reward for cooperative multiagent reinforcement learning |
| title_short | Generating individual intrinsic reward for cooperative multiagent reinforcement learning |
| title_sort | generating individual intrinsic reward for cooperative multiagent reinforcement learning |
| url | https://doi.org/10.1177/17298814211044946 |
| work_keys_str_mv | AT haolinwu generatingindividualintrinsicrewardforcooperativemultiagentreinforcementlearning AT huili generatingindividualintrinsicrewardforcooperativemultiagentreinforcementlearning AT jianweizhang generatingindividualintrinsicrewardforcooperativemultiagentreinforcementlearning AT zhuangwang generatingindividualintrinsicrewardforcooperativemultiagentreinforcementlearning AT jianengzhang generatingindividualintrinsicrewardforcooperativemultiagentreinforcementlearning |