One-shot learning by inverting a compositional causal process
People can learn a new visual class from just one example, yet machine learning algorithms typically require hundreds or thousands of examples to tackle the same problems. Here we present a Hierarchical Bayesian model based on compositionality and causality that can learn a wide range of natural (al...
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | en_US |
| Published: |
Neural Information Processing Systems Foundation, Inc.
2015
|
| Online Access: | http://hdl.handle.net/1721.1/94624 https://orcid.org/0000-0002-1925-2035 |
| _version_ | 1826194122356031488 |
|---|---|
| author | Lake, Brenden M. Salakhutdinov, Ruslan Tenenbaum, Joshua B. |
| author2 | Massachusetts Institute of Technology. Department of Brain and Cognitive Sciences |
| author_facet | Massachusetts Institute of Technology. Department of Brain and Cognitive Sciences Lake, Brenden M. Salakhutdinov, Ruslan Tenenbaum, Joshua B. |
| author_sort | Lake, Brenden M. |
| collection | MIT |
| description | People can learn a new visual class from just one example, yet machine learning algorithms typically require hundreds or thousands of examples to tackle the same problems. Here we present a Hierarchical Bayesian model based on compositionality and causality that can learn a wide range of natural (although simple) visual concepts, generalizing in human-like ways from just one image. We evaluated performance on a challenging one-shot classification task, where our model achieved a human-level error rate while substantially outperforming two deep learning models. We also used a visual Turing test "to show that our model produces human-like performance on other conceptual tasks, including generating new examples and parsing." |
| first_indexed | 2024-09-23T09:50:32Z |
| format | Article |
| id | mit-1721.1/94624 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T09:50:32Z |
| publishDate | 2015 |
| publisher | Neural Information Processing Systems Foundation, Inc. |
| record_format | dspace |
| spelling | mit-1721.1/946242022-09-26T14:04:34Z One-shot learning by inverting a compositional causal process Lake, Brenden M. Salakhutdinov, Ruslan Tenenbaum, Joshua B. Massachusetts Institute of Technology. Department of Brain and Cognitive Sciences Lake, Brenden M. Tenenbaum, Joshua B. People can learn a new visual class from just one example, yet machine learning algorithms typically require hundreds or thousands of examples to tackle the same problems. Here we present a Hierarchical Bayesian model based on compositionality and causality that can learn a wide range of natural (although simple) visual concepts, generalizing in human-like ways from just one image. We evaluated performance on a challenging one-shot classification task, where our model achieved a human-level error rate while substantially outperforming two deep learning models. We also used a visual Turing test "to show that our model produces human-like performance on other conceptual tasks, including generating new examples and parsing." National Science Foundation (U.S.) (NSF Graduate Research Fellowship) United States. Army Research Office (ARO MURI contract W911NF-08-1-0242) 2015-02-18T21:30:15Z 2015-02-18T21:30:15Z 2013-12 Article http://purl.org/eprint/type/ConferencePaper http://hdl.handle.net/1721.1/94624 Lake, Brenden M., Ruslan Salakhutdinov and Joshua B. Tenenbaum. "One-shot learning by inverting a compositional causal process." Advances in Neural Information Processing Systems 26, NIPS 2013, Lake Tahoe, Nevada, United States, December 5-10, 2013. https://orcid.org/0000-0002-1925-2035 en_US http://papers.nips.cc/paper/5128-one-shot-learning-by-inverting-a-compositional-causal-process Advances in Neural Information Processing Systems 26 (NIPS 2013) Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf Neural Information Processing Systems Foundation, Inc. University of Toronoto web domain |
| spellingShingle | Lake, Brenden M. Salakhutdinov, Ruslan Tenenbaum, Joshua B. One-shot learning by inverting a compositional causal process |
| title | One-shot learning by inverting a compositional causal process |
| title_full | One-shot learning by inverting a compositional causal process |
| title_fullStr | One-shot learning by inverting a compositional causal process |
| title_full_unstemmed | One-shot learning by inverting a compositional causal process |
| title_short | One-shot learning by inverting a compositional causal process |
| title_sort | one shot learning by inverting a compositional causal process |
| url | http://hdl.handle.net/1721.1/94624 https://orcid.org/0000-0002-1925-2035 |
| work_keys_str_mv | AT lakebrendenm oneshotlearningbyinvertingacompositionalcausalprocess AT salakhutdinovruslan oneshotlearningbyinvertingacompositionalcausalprocess AT tenenbaumjoshuab oneshotlearningbyinvertingacompositionalcausalprocess |