{"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/towards-universal-neural-nets-gibbs-machines","title":"Towards universal neural nets: Gibbs machines and ACE","arxiv_id":"1508.06585","date":"2015-08-26","proceeding":null,"authors":["Galin Georgiev"],"abstract":"We study from a physics viewpoint a class of generative neural nets, Gibbs\nmachines, designed for gradual learning. While including variational\nauto-encoders, they offer a broader universal platform for incrementally adding\nnewly learned features, including physical symmetries. Their direct connection\nto statistical physics and information geometry is established. A variational\nPythagorean theorem justifies invoking the exponential/Gibbs class of\nprobabilities for creating brand new objects. Combining these nets with\nclassifiers, gives rise to a brand of universal generative neural nets -\nstochastic auto-classifier-encoders (ACE). ACE have state-of-the-art\nperformance in their class, both for classification and density estimation for\nthe MNIST data set.","url_abs":"http://arxiv.org/abs/1508.06585v5","url_pdf":"http://arxiv.org/pdf/1508.06585v5.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":"towards-universal-neural-nets-gibbs-machines","repo_url":"https://github.com/galinngeorgiev/ACE","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"density-estimation","task_name":"Density Estimation"},{"task_slug":"classification","task_name":"General Classification"}],"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}