{"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/efficient-probabilistic-inference-in-generic","title":"Efficient Probabilistic Inference in Generic Neural Networks Trained with Non-Probabilistic Feedback","arxiv_id":"1601.03060","date":"2016-01-12","proceeding":null,"authors":["A. Emin Orhan","Wei Ji Ma"],"abstract":"Animals perform near-optimal probabilistic inference in a wide range of\npsychophysical tasks. Probabilistic inference requires trial-to-trial\nrepresentation of the uncertainties associated with task variables and\nsubsequent use of this representation. Previous work has implemented such\ncomputations using neural networks with hand-crafted and task-dependent\noperations. We show that generic neural networks trained with a simple\nerror-based learning rule perform near-optimal probabilistic inference in nine\ncommon psychophysical tasks. In a probabilistic categorization task,\nerror-based learning in a generic network simultaneously explains a monkey's\nlearning curve and the evolution of qualitative aspects of its choice behavior.\nIn all tasks, the number of neurons required for a given level of performance\ngrows sub-linearly with the input population size, a substantial improvement on\nprevious implementations of probabilistic inference. The trained networks\ndevelop a novel sparsity-based probabilistic population code. Our results\nsuggest that probabilistic inference emerges naturally in generic neural\nnetworks trained with error-based learning rules.","url_abs":"http://arxiv.org/abs/1601.03060v4","url_pdf":"http://arxiv.org/pdf/1601.03060v4.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":"efficient-probabilistic-inference-in-generic","repo_url":"https://github.com/eminorhan/inevitable-probability","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}