File(s) under permanent embargo
Hierarchical and non-hierarchical multi-agent interactions based on unity reinforcement learning
The open-source Unity platform, where agents can be trained using hierarchical or non-hierarchical reinforcement learning, supports the use of games and simulations as environments for multipleagent interactions. In this demonstration, we present hierarchical and non-hierarchical multi-agent interactions based on Unity reinforcement learning, specifically, hierarchical reinforcement learning that sets different levels of agent’s observations to achieve the goal. We created four multi-agent scenarios in the Unity environment, namely, Crawler, Tennis, Banana Collector, and Soccer, to test the interaction performances of hierarchical and nonhierarchical reinforcement learning. The simulation-interaction performances show that hierarchical reinforcement learning can be applied to multi-agent environments and can compete with agents trained via non-hierarchical reinforcement learning.
The demonstration video can be viewed at the following link: https://youtu.be/YQYQwLPXaL4
History
Publication title
Proceedings of the 19th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2020)Editors
B An, N Yorke-Smith, A El Fallah Seghrouchni, and G SukthankarPagination
2095-2097Department/School
School of Information and Communication TechnologyPublisher
International Foundation for Autonomous Agents and Multiagent SystemsPlace of publication
United StatesEvent title
19th International Conference on Autonomous Agents and Multiagent Systems 2020Event Venue
University of Auckland (virtual/online)Date of Event (Start Date)
2020-05-09Date of Event (End Date)
2020-05-13Rights statement
Copyright 2020 International Foundation for Autonomous Agents and Multiagent SystemsRepository Status
- Restricted