{"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/connectionist-learning-of-belief-networks","title":"Connectionist Learning of Belief Networks","arxiv_id":null,"date":"1992-06-01","proceeding":null,"authors":["Radford M. Neal"],"abstract":"Connectionist learning procedures are presented for \"sigmoid\" and \"noisy-OR\" varieties\r\nof probabilistic belief networks. These networks have previously been seen primarily as a\r\nmeans of representing knowledge derived from experts. Here it is shown that the \"Gibbs\r\nsampling\" simulation procedure for such networks can support maximum-likelihood\r\nlearning from empirical data through local gradient ascent. This learning procedure\r\nresembles that used for \"Boltzmann machines\", and like it, allows the use of \"hidden\"\r\nvariables to model correlations between visible variables. Due to the directed nature\r\nof the connections in a belief network, however, the \"negative phase\" of Boltzmann\r\nmachine learning is unnecessary. Experimental results show that, as a result, learning in\r\na sigmoid belief network can be faster than in a Boltzmann machine. These networks\r\nhave other advantages over Boltzmann machines in pattern classification and decision\r\nmaking applications, are naturally applicable to unsupervised learning problems, and\r\nprovide a link between work on connectionist learning and work on the representation\r\nof expert knowledge.","url_abs":"https://www.semanticscholar.org/paper/Connectionist-Learning-of-Belief-Networks-Neal/a120c05ad7cd4ce2eb8fb9697e16c7c4877208a5","url_pdf":"http://www.cs.toronto.edu/~bonner/courses/2016s/csc321/readings/Connectionist%20learning%20of%20belief%20networks.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":"connectionist-learning-of-belief-networks","repo_url":"https://github.com/EugenHotaj/pytorch-generative/blob/master/pytorch_generative/models/autoregressive/fvbn.py","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"decision-making","task_name":"Decision Making"}],"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}