{"id":190870,"date":"2014-05-15T00:00:00","date_gmt":"2014-05-15T06:10:55","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/msr-research-item\/principled-approaches-for-learning-latent-variable-models\/"},"modified":"2016-07-15T15:17:04","modified_gmt":"2016-07-15T22:17:04","slug":"principled-approaches-for-learning-latent-variable-models","status":"publish","type":"msr-video","link":"https:\/\/www.microsoft.com\/en-us\/research\/video\/principled-approaches-for-learning-latent-variable-models\/","title":{"rendered":"Principled Approaches for Learning Latent Variable Models"},"content":{"rendered":"<div class=\"asset-content\">\n<p>In any learning task, it is natural to incorporate latent or hidden variables which are not directly observed. For instance, in a social network, we can observe interactions among the actors, but not their hidden interests\/intents, in gene networks, we can  measure gene expression levels but not the detailed regulatory mechanisms, and so on. I will present a broad framework for unsupervised learning of latent variable models, addressing both statistical and computational concerns. We show that higher order relationships among observed variables have a low rank representation under  natural statistical constraints such as conditional-independence relationships. We also present efficient computational methods for finding these low rank representations. These findings have implications in a number of settings such as finding hidden communities in networks, discovering topics in text documents and learning about gene regulation in computational biology. I will also present principled approaches for learning overcomplete models, where the latent dimensionality can be much larger than the observed dimensionality, under natural sparsity constraints. This has implications in a number of applications such as sparse coding and feature learning.<\/p>\n<\/div>\n<p><!-- .asset-content --><\/p>\n","protected":false},"excerpt":{"rendered":"<p>In any learning task, it is natural to incorporate latent or hidden variables which are not directly observed. For instance, in a social network, we can observe interactions among the actors, but not their hidden interests\/intents, in gene networks, we can measure gene expression levels but not the detailed regulatory mechanisms, and so on. I [&hellip;]<\/p>\n","protected":false},"featured_media":198347,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","footnotes":""},"research-area":[],"msr-video-type":[],"msr-locale":[268875],"msr-post-option":[],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-190870","msr-video","type-msr-video","status-publish","has-post-thumbnail","hentry","msr-locale-en_us"],"msr_download_urls":"","msr_external_url":"https:\/\/youtu.be\/SAa_1YiGH8c","msr_secondary_video_url":"","msr_video_file":"","_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-video\/190870"}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-video"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-video"}],"version-history":[{"count":0,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-video\/190870\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media\/198347"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=190870"}],"wp:term":[{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=190870"},{"taxonomy":"msr-video-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-video-type?post=190870"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=190870"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=190870"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=190870"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=190870"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}