{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/learning-physical-intuition-of-block-towers","title":"Learning Physical Intuition of Block Towers by Example","arxiv_id":"1603.01312","date":"2016-03-03","proceeding":null,"authors":["Adam Lerer","Sam Gross","Rob Fergus"],"abstract":"Wooden blocks are a common toy for infants, allowing them to develop motor\nskills and gain intuition about the physical behavior of the world. In this\npaper, we explore the ability of deep feed-forward models to learn such\nintuitive physics. Using a 3D game engine, we create small towers of wooden\nblocks whose stability is randomized and render them collapsing (or remaining\nupright). This data allows us to train large convolutional network models which\ncan accurately predict the outcome, as well as estimating the block\ntrajectories. The models are also able to generalize in two important ways: (i)\nto new physical scenarios, e.g. towers with an additional block and (ii) to\nimages of real wooden blocks, where it obtains a performance comparable to\nhuman subjects.","url_abs":"http://arxiv.org/abs/1603.01312v1","url_pdf":"http://arxiv.org/pdf/1603.01312v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"learning-physical-intuition-of-block-towers","repo_url":"https://github.com/facebook/UETorch","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"torch","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"learning-physical-intuition-of-block-towers","repo_url":"https://github.com/facebookarchive/uetorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"torch","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"learning-physical-intuition-of-block-towers","repo_url":"https://github.com/ogroth/shapestacks","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"GPL-3.0"}}],"tasks":[{"task_slug":"physical-intuition","task_name":"Physical Intuition"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1603.01312","atlas_url":"https://app.syntology.ai/?focus=1603.01312","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}