{"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/distilling-a-neural-network-into-a-soft","title":"Distilling a Neural Network Into a Soft Decision Tree","arxiv_id":"1711.09784","date":"2017-11-27","proceeding":null,"authors":["Nicholas Frosst","Geoffrey Hinton"],"abstract":"Deep neural networks have proved to be a very effective way to perform\nclassification tasks. They excel when the input data is high dimensional, the\nrelationship between the input and the output is complicated, and the number of\nlabeled training examples is large. But it is hard to explain why a learned\nnetwork makes a particular classification decision on a particular test case.\nThis is due to their reliance on distributed hierarchical representations. If\nwe could take the knowledge acquired by the neural net and express the same\nknowledge in a model that relies on hierarchical decisions instead, explaining\na particular decision would be much easier. We describe a way of using a\ntrained neural net to create a type of soft decision tree that generalizes\nbetter than one learned directly from the training data.","url_abs":"http://arxiv.org/abs/1711.09784v1","url_pdf":"http://arxiv.org/pdf/1711.09784v1.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":"distilling-a-neural-network-into-a-soft","repo_url":"https://github.com/AaronX121/Soft-Decision-Tree","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"distilling-a-neural-network-into-a-soft","repo_url":"https://github.com/kimhc6028/soft-decision-tree","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"distilling-a-neural-network-into-a-soft","repo_url":"https://github.com/lmartak/distill-nn-tree","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"distilling-a-neural-network-into-a-soft","repo_url":"https://github.com/pinkTheRock/Explainable-Deep-Learning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"distilling-a-neural-network-into-a-soft","repo_url":"https://github.com/xuyxu/Soft-Decision-Tree","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"distilling-a-neural-network-into-a-soft","repo_url":"https://github.com/endymion64/SoftDecisionTree","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"classification","task_name":"General Classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1711.09784","atlas_url":"https://app.syntology.ai/?focus=1711.09784","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}