{"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/end-to-end-learning-of-deterministic-decision","title":"End-to-end Learning of Deterministic Decision Trees","arxiv_id":"1712.02743","date":"2017-12-07","proceeding":null,"authors":["Thomas Hehn","Fred A. Hamprecht"],"abstract":"Conventional decision trees have a number of favorable properties, including\ninterpretability, a small computational footprint and the ability to learn from\nlittle training data. However, they lack a key quality that has helped fuel the\ndeep learning revolution: that of being end-to-end trainable, and to learn from\nscratch those features that best allow to solve a given supervised learning\nproblem. Recent work (Kontschieder 2015) has addressed this deficit, but at the\ncost of losing a main attractive trait of decision trees: the fact that each\nsample is routed along a small subset of tree nodes only. We here propose a\nmodel and Expectation-Maximization training scheme for decision trees that are\nfully probabilistic at train time, but after a deterministic annealing process\nbecome deterministic at test time. We also analyze the learned oblique split\nparameters on image datasets and show that Neural Networks can be trained at\neach split node. In summary, we present the first end-to-end learning scheme\nfor deterministic decision trees and present results on par with or superior to\npublished standard oblique decision tree algorithms.","url_abs":"http://arxiv.org/abs/1712.02743v1","url_pdf":"http://arxiv.org/pdf/1712.02743v1.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":"end-to-end-learning-of-deterministic-decision","repo_url":"https://github.com/tomsal/endtoenddecisiontrees","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}