Human learning in Atari
Atari games are an excellent testbed for studying intelligent behavior, as they offer a range of tasks that differ widely in their visual representation, game dynamics, and goals presented to an agent. The last two years have seen a spate of research into artificial agents that use a single algorith...
| Main Authors: | , , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Published: |
Association for the Advancement of Artificial Intelligence
2017
|
| Online Access: | http://hdl.handle.net/1721.1/112620 https://orcid.org/0000-0002-0138-163X https://orcid.org/0000-0002-1925-2035 |
| _version_ | 1826217359956770816 |
|---|---|
| author | Pouncy, Thomas Gershman, Samuel J. Tsividis, Pedro Xu, Jacqueline L. Tenenbaum, Joshua B |
| author2 | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
| author_facet | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Pouncy, Thomas Gershman, Samuel J. Tsividis, Pedro Xu, Jacqueline L. Tenenbaum, Joshua B |
| author_sort | Pouncy, Thomas |
| collection | MIT |
| description | Atari games are an excellent testbed for studying intelligent behavior, as they offer a range of tasks that differ widely in their visual representation, game dynamics, and goals presented to an agent. The last two years have seen a spate of research into artificial agents that use a single algorithm to learn to play these games. The best of these artificial agents perform at better-than-human levels on most games, but require hundreds of hours of game-play experience to produce such behavior. Humans, on the other hand, can learn to perform well on these tasks in a matter of minutes. In this paper we present data on human learning trajectories for several Atari games, and test several hypotheses about the mechanisms that lead to such rapid learning. |
| first_indexed | 2024-09-23T17:02:24Z |
| format | Article |
| id | mit-1721.1/112620 |
| institution | Massachusetts Institute of Technology |
| last_indexed | 2024-09-23T17:02:24Z |
| publishDate | 2017 |
| publisher | Association for the Advancement of Artificial Intelligence |
| record_format | dspace |
| spelling | mit-1721.1/1126202022-09-29T23:16:46Z Human learning in Atari Pouncy, Thomas Gershman, Samuel J. Tsividis, Pedro Xu, Jacqueline L. Tenenbaum, Joshua B Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Massachusetts Institute of Technology. Department of Brain and Cognitive Sciences Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Tsividis, Pedro Xu, Jacqueline L. Tenenbaum, Joshua B Atari games are an excellent testbed for studying intelligent behavior, as they offer a range of tasks that differ widely in their visual representation, game dynamics, and goals presented to an agent. The last two years have seen a spate of research into artificial agents that use a single algorithm to learn to play these games. The best of these artificial agents perform at better-than-human levels on most games, but require hundreds of hours of game-play experience to produce such behavior. Humans, on the other hand, can learn to perform well on these tasks in a matter of minutes. In this paper we present data on human learning trajectories for several Atari games, and test several hypotheses about the mechanisms that lead to such rapid learning. National Science Foundation (U.S.) (Award CCF-1231216) 2017-12-07T15:30:24Z 2017-12-07T15:30:24Z 2017 2017-12-06T14:38:51Z Article http://purl.org/eprint/type/ConferencePaper http://hdl.handle.net/1721.1/112620 Tsividis, Pedro A. et al. "Human learning in Atari." 2017 AAAI Spring Symposium Series, Science of Intelligence: Computational Principles of Natural and Artificial Intelligence, Technical Report SS-17-07 (2017) © 2017 Association for the Advancement of Artificial Intelligence https://orcid.org/0000-0002-0138-163X https://orcid.org/0000-0002-1925-2035 https://aaai.org/ocs/index.php/SSS/SSS17/paper/viewFile/15280/14616 2017 AAAI Spring Symposium Series, Science of Intelligence: Computational Principles of Natural and Artificial Intelligence Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf Association for the Advancement of Artificial Intelligence Other univ. web domain |
| spellingShingle | Pouncy, Thomas Gershman, Samuel J. Tsividis, Pedro Xu, Jacqueline L. Tenenbaum, Joshua B Human learning in Atari |
| title | Human learning in Atari |
| title_full | Human learning in Atari |
| title_fullStr | Human learning in Atari |
| title_full_unstemmed | Human learning in Atari |
| title_short | Human learning in Atari |
| title_sort | human learning in atari |
| url | http://hdl.handle.net/1721.1/112620 https://orcid.org/0000-0002-0138-163X https://orcid.org/0000-0002-1925-2035 |
| work_keys_str_mv | AT pouncythomas humanlearninginatari AT gershmansamuelj humanlearninginatari AT tsividispedro humanlearninginatari AT xujacquelinel humanlearninginatari AT tenenbaumjoshuab humanlearninginatari |