Human-level performance in 3D multiplayer games with population-based reinforcement learning.
Saved in:
| Title: | Human-level performance in 3D multiplayer games with population-based reinforcement learning. |
|---|---|
| Authors: | Jaderberg, Max (AUTHOR), Czarnecki, Wojciech M. (AUTHOR), Dunning, Iain (AUTHOR), Marris, Luke (AUTHOR), Lever, Guy (AUTHOR), Castañeda, Antonio Garcia (AUTHOR), Beattie, Charles (AUTHOR), Rabinowitz, Neil C. (AUTHOR), Morcos, Ari S. (AUTHOR), Ruderman, Avraham (AUTHOR), Sonnerat, Nicolas (AUTHOR), Green, Tim (AUTHOR), Deason, Louise (AUTHOR), Leibo, Joel Z. (AUTHOR), Silver, David (AUTHOR), Hassabis, Demis (AUTHOR), Kavukcuoglu, Koray (AUTHOR), Graepel, Thore (AUTHOR) |
| Source: | Science (pre-March 2025). 5/31/2019, Vol. 364 Issue 6443, p859-865. 7p. 4 Diagrams. |
| Subjects: | Reinforcement learning, Multiplayer games, Agent (Philosophy), Pixels, Multiagent systems, Artificial intelligence research, Video games |
| Abstract: | Reinforcement learning (RL) has shown great success in increasingly complex single-agent environments and two-player turn-based games. However, the real world contains multiple agents, each learning and acting independently to cooperate and compete with other agents. We used a tournament-style evaluation to demonstrate that an agent can achieve human-level performance in a three-dimensional multiplayer first-person video game, Quake III Arena in Capture the Flag mode, using only pixels and game points scored as input. We used a two-tier optimization process in which a population of independent RL agents are trained concurrently from thousands of parallel matches on randomly generated environments. Each agent learns its own internal reward signal and rich representation of the world. These results indicate the great potential of multiagent reinforcement learning for artificial intelligence research. [ABSTRACT FROM AUTHOR] |
| Copyright of Science (pre-March 2025) is the property of American Association for the Advancement of Science and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Psychology and Behavioral Sciences Collection |
|
Full text is not displayed to guests.
Login for full access.
|
|
| Abstract: | Reinforcement learning (RL) has shown great success in increasingly complex single-agent environments and two-player turn-based games. However, the real world contains multiple agents, each learning and acting independently to cooperate and compete with other agents. We used a tournament-style evaluation to demonstrate that an agent can achieve human-level performance in a three-dimensional multiplayer first-person video game, Quake III Arena in Capture the Flag mode, using only pixels and game points scored as input. We used a two-tier optimization process in which a population of independent RL agents are trained concurrently from thousands of parallel matches on randomly generated environments. Each agent learns its own internal reward signal and rich representation of the world. These results indicate the great potential of multiagent reinforcement learning for artificial intelligence research. [ABSTRACT FROM AUTHOR] |
|---|---|
| ISSN: | 00368075 |
| DOI: | 10.1126/science.aau6249 |
