{"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/quantifying-the-effects-of-enforcing","title":"Quantifying the Effects of Enforcing Disentanglement on Variational Autoencoders","arxiv_id":"1711.09159","date":"2017-11-24","proceeding":null,"authors":["Momchil Peychev","Petar Veličković","Pietro Liò"],"abstract":"The notion of disentangled autoencoders was proposed as an extension to the\nvariational autoencoder by introducing a disentanglement parameter $\\beta$,\ncontrolling the learning pressure put on the possible underlying latent\nrepresentations. For certain values of $\\beta$ this kind of autoencoders is\ncapable of encoding independent input generative factors in separate elements\nof the code, leading to a more interpretable and predictable model behaviour.\nIn this paper we quantify the effects of the parameter $\\beta$ on the model\nperformance and disentanglement. After training multiple models with the same\nvalue of $\\beta$, we establish the existence of consistent variance in one of\nthe disentanglement measures, proposed in literature. The negative consequences\nof the disentanglement to the autoencoder's discriminative ability are also\nasserted while varying the amount of examples available during training.","url_abs":"http://arxiv.org/abs/1711.09159v1","url_pdf":"http://arxiv.org/pdf/1711.09159v1.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":"quantifying-the-effects-of-enforcing","repo_url":"https://github.com/mpeychev/disentangled-autoencoders","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"disentanglement","task_name":"Disentanglement"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}