From Few to More: Large-Scale Dynamic Multiagent Curriculum Learning
العنوان: | From Few to More: Large-Scale Dynamic Multiagent Curriculum Learning |
---|---|
المؤلفون: | Yong Liu, Changjie Fan, Yujing Hu, Weixun Wang, Jianye Hao, Yang Gao, Tianpei Yang, Xiaotian Hao, Yingfeng Chen |
المصدر: | AAAI |
بيانات النشر: | Association for the Advancement of Artificial Intelligence (AAAI), 2020. |
سنة النشر: | 2020 |
مصطلحات موضوعية: | FOS: Computer and information sciences, Computer Science - Artificial Intelligence, Computer science, Process (engineering), Distributed computing, Scale (chemistry), Multi-agent system, 02 engineering and technology, General Medicine, Complex dynamics, Artificial Intelligence (cs.AI), 020204 information systems, 0202 electrical engineering, electronic engineering, information engineering, Reinforcement learning, Computer Science - Multiagent Systems, 020201 artificial intelligence & image processing, State (computer science), Dimension (data warehouse), Curriculum, Multiagent Systems (cs.MA) |
الوصف: | A lot of efforts have been devoted to investigating how agents can learn effectively and achieve coordination in multiagent systems. However, it is still challenging in large-scale multiagent settings due to the complex dynamics between the environment and agents and the explosion of state-action space. In this paper, we design a novel Dynamic Multiagent Curriculum Learning (DyMA-CL) to solve large-scale problems by starting from learning on a multiagent scenario with a small size and progressively increasing the number of agents. We propose three transfer mechanisms across curricula to accelerate the learning process. Moreover, due to the fact that the state dimension varies across curricula,, and existing network structures cannot be applied in such a transfer setting since their network input sizes are fixed. Therefore, we design a novel network structure called Dynamic Agent-number Network (DyAN) to handle the dynamic size of the network input. Experimental results show that DyMA-CL using DyAN greatly improves the performance of large-scale multiagent learning compared with state-of-the-art deep reinforcement learning approaches. We also investigate the influence of three transfer mechanisms across curricula through extensive simulations. Comment: Accepted by AAAI2020 |
تدمد: | 2374-3468 2159-5399 |
URL الوصول: | https://explore.openaire.eu/search/publication?articleId=doi_dedup___::7346708eb2b39caa237d87d6d2cedd2f https://doi.org/10.1609/aaai.v34i05.6221 |
حقوق: | OPEN |
رقم الأكسشن: | edsair.doi.dedup.....7346708eb2b39caa237d87d6d2cedd2f |
قاعدة البيانات: | OpenAIRE |
تدمد: | 23743468 21595399 |
---|