{"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/learning-disentangled-representations-with-1","title":"Learning Disentangled Representations with Semi-Supervised Deep Generative Models","arxiv_id":"1706.00400","date":"2017-06-01","proceeding":"NeurIPS 2017 12","authors":["N. Siddharth","Brooks Paige","Jan-Willem van de Meent","Alban Desmaison","Noah D. Goodman","Pushmeet Kohli","Frank Wood","Philip H. S. Torr"],"abstract":"Variational autoencoders (VAEs) learn representations of data by jointly\ntraining a probabilistic encoder and decoder network. Typically these models\nencode all features of the data into a single variable. Here we are interested\nin learning disentangled representations that encode distinct aspects of the\ndata into separate variables. We propose to learn such representations using\nmodel architectures that generalise from standard VAEs, employing a general\ngraphical model structure in the encoder and decoder. This allows us to train\npartially-specified models that make relatively strong assumptions about a\nsubset of interpretable variables and rely on the flexibility of neural\nnetworks to learn representations for the remaining variables. We further\ndefine a general objective for semi-supervised learning in this model class,\nwhich can be approximated using an importance sampling procedure. We evaluate\nour framework's ability to learn disentangled representations, both by\nqualitative exploration of its generative capacity, and quantitative evaluation\nof its discriminative ability on a variety of models and datasets.","url_abs":"http://arxiv.org/abs/1706.00400v2","url_pdf":"http://arxiv.org/pdf/1706.00400v2.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":"learning-disentangled-representations-with-1","repo_url":"https://github.com/probtorch/probtorch","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"representation-learning","task_name":"Representation Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1706.00400","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}