{"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/stochastic-generative-hashing","title":"Stochastic Generative Hashing","arxiv_id":"1701.02815","date":"2017-01-11","proceeding":"ICML 2017 8","authors":["Bo Dai","Ruiqi Guo","Sanjiv Kumar","Niao He","Le Song"],"abstract":"Learning-based binary hashing has become a powerful paradigm for fast search\nand retrieval in massive databases. However, due to the requirement of discrete\noutputs for the hash functions, learning such functions is known to be very\nchallenging. In addition, the objective functions adopted by existing hashing\ntechniques are mostly chosen heuristically. In this paper, we propose a novel\ngenerative approach to learn hash functions through Minimum Description Length\nprinciple such that the learned hash codes maximally compress the dataset and\ncan also be used to regenerate the inputs. We also develop an efficient\nlearning algorithm based on the stochastic distributional gradient, which\navoids the notorious difficulty caused by binary output constraints, to jointly\noptimize the parameters of the hash function and the associated generative\nmodel. Extensive experiments on a variety of large-scale datasets show that the\nproposed method achieves better retrieval results than the existing\nstate-of-the-art methods.","url_abs":"http://arxiv.org/abs/1701.02815v2","url_pdf":"http://arxiv.org/pdf/1701.02815v2.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":"stochastic-generative-hashing","repo_url":"https://github.com/doubling/Stochastic_Generative_Hashing","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"stochastic-generative-hashing","repo_url":"https://github.com/zhangcheng-007/Stochastic_Generative_Hashing","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"retrieval","task_name":"Retrieval"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1701.02815","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}