{"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/black-box-divergence-minimization","title":"Black-box $α$-divergence Minimization","arxiv_id":"1511.03243","date":"2015-11-10","proceeding":null,"authors":["José Miguel Hernández-Lobato","Yingzhen Li","Mark Rowland","Daniel Hernández-Lobato","Thang Bui","Richard E. Turner"],"abstract":"Black-box alpha (BB-$\\alpha$) is a new approximate inference method based on\nthe minimization of $\\alpha$-divergences. BB-$\\alpha$ scales to large datasets\nbecause it can be implemented using stochastic gradient descent. BB-$\\alpha$\ncan be applied to complex probabilistic models with little effort since it only\nrequires as input the likelihood function and its gradients. These gradients\ncan be easily obtained using automatic differentiation. By changing the\ndivergence parameter $\\alpha$, the method is able to interpolate between\nvariational Bayes (VB) ($\\alpha \\rightarrow 0$) and an algorithm similar to\nexpectation propagation (EP) ($\\alpha = 1$). Experiments on probit regression\nand neural network regression and classification problems show that BB-$\\alpha$\nwith non-standard settings of $\\alpha$, such as $\\alpha = 0.5$, usually\nproduces better predictions than with $\\alpha \\rightarrow 0$ (VB) or $\\alpha =\n1$ (EP).","url_abs":"http://arxiv.org/abs/1511.03243v3","url_pdf":"http://arxiv.org/pdf/1511.03243v3.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":"black-box-divergence-minimization","repo_url":"https://github.com/SuperKam91/bnn","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"black-box-divergence-minimization","repo_url":"https://github.com/tensorflow/models","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"black-box-divergence-minimization","repo_url":"https://github.com/tensorflow/models/tree/master/research/deep_contextual_bandits","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"regression-1","task_name":"regression"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1511.03243","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}