Building machines that learn and think like people
Recent progress in artificial intelligence (AI) has renewed interest in building systems that learn and think like people. Many advances have come from using deep neural networks trained end-to-end in tasks such as object recognition, video games, and board games, achieving performance that equals o...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Published: |
Cambridge University Press
2017
|
| Online Access: | http://hdl.handle.net/1721.1/112658 https://orcid.org/0000-0003-1722-2382 https://orcid.org/0000-0002-1925-2035 |
| _version_ | 1826217442010988544 |
|---|---|
| author | Lake, Brenden M. Ullman, Tomer David Tenenbaum, Joshua B Gershman, Samuel J |
| author2 | Center for Brains, Minds, and Machines |
| author_facet | Center for Brains, Minds, and Machines Lake, Brenden M. Ullman, Tomer David Tenenbaum, Joshua B Gershman, Samuel J |
| author_sort | Lake, Brenden M. |
| collection | MIT |
| description | Recent progress in artificial intelligence (AI) has renewed interest in building systems that learn and think like people. Many advances have come from using deep neural networks trained end-to-end in tasks such as object recognition, video games, and board games, achieving performance that equals or even beats humans in some respects. Despite their biological inspiration and performance achievements, these systems differ from human intelligence in crucial ways. We review progress in cognitive science suggesting that truly human-like learning and thinking machines will have to reach beyond current engineering trends in both what they learn, and how they learn it. Specifically, we argue that these machines should (a) build causal models of the world that support explanation and understanding, rather than merely solving pattern recognition problems; (b) ground learning in intuitive theories of physics and psychology, to support and enrich the knowledge that is learned; and (c) harness compositionality and learning-to-learn to rapidly acquire and generalize knowledge to new tasks and situations. We suggest concrete challenges and promising routes towards these goals that can combine the strengths of recent neural network advances with more structured cognitive models. |
| first_indexed | 2024-09-23T17:03:43Z |
| format | Article |
| id | mit-1721.1/112658 |
| institution | Massachusetts Institute of Technology |
| last_indexed | 2024-09-23T17:03:43Z |
| publishDate | 2017 |
| publisher | Cambridge University Press |
| record_format | dspace |
| spelling | mit-1721.1/1126582022-10-03T10:07:20Z Building machines that learn and think like people Lake, Brenden M. Ullman, Tomer David Tenenbaum, Joshua B Gershman, Samuel J Center for Brains, Minds, and Machines Massachusetts Institute of Technology. Department of Brain and Cognitive Sciences Ullman, Tomer David Tenenbaum, Joshua B Gershman, Samuel J Recent progress in artificial intelligence (AI) has renewed interest in building systems that learn and think like people. Many advances have come from using deep neural networks trained end-to-end in tasks such as object recognition, video games, and board games, achieving performance that equals or even beats humans in some respects. Despite their biological inspiration and performance achievements, these systems differ from human intelligence in crucial ways. We review progress in cognitive science suggesting that truly human-like learning and thinking machines will have to reach beyond current engineering trends in both what they learn, and how they learn it. Specifically, we argue that these machines should (a) build causal models of the world that support explanation and understanding, rather than merely solving pattern recognition problems; (b) ground learning in intuitive theories of physics and psychology, to support and enrich the knowledge that is learned; and (c) harness compositionality and learning-to-learn to rapidly acquire and generalize knowledge to new tasks and situations. We suggest concrete challenges and promising routes towards these goals that can combine the strengths of recent neural network advances with more structured cognitive models. National Science Foundation (U.S.) (Award CCF-1231216) 2017-12-08T16:29:31Z 2017-12-08T16:29:31Z 2016-11 2017-12-08T14:42:47Z Article http://purl.org/eprint/type/JournalArticle 0140-525X 1469-1825 http://hdl.handle.net/1721.1/112658 Lake, Brenden M. et al. “Building Machines That Learn and Think Like People.” Behavioral and Brain Sciences 40 (November 2016): e253 © 2016 Cambridge University Press https://orcid.org/0000-0003-1722-2382 https://orcid.org/0000-0002-1925-2035 http://dx.doi.org/10.1017/S0140525X16001837 Behavioral and Brain Sciences Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf Cambridge University Press arXiv |
| spellingShingle | Lake, Brenden M. Ullman, Tomer David Tenenbaum, Joshua B Gershman, Samuel J Building machines that learn and think like people |
| title | Building machines that learn and think like people |
| title_full | Building machines that learn and think like people |
| title_fullStr | Building machines that learn and think like people |
| title_full_unstemmed | Building machines that learn and think like people |
| title_short | Building machines that learn and think like people |
| title_sort | building machines that learn and think like people |
| url | http://hdl.handle.net/1721.1/112658 https://orcid.org/0000-0003-1722-2382 https://orcid.org/0000-0002-1925-2035 |
| work_keys_str_mv | AT lakebrendenm buildingmachinesthatlearnandthinklikepeople AT ullmantomerdavid buildingmachinesthatlearnandthinklikepeople AT tenenbaumjoshuab buildingmachinesthatlearnandthinklikepeople AT gershmansamuelj buildingmachinesthatlearnandthinklikepeople |