{"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/taming-vaes","title":"Taming VAEs","arxiv_id":"1810.00597","date":"2018-10-01","proceeding":null,"authors":["Danilo Jimenez Rezende","Fabio Viola"],"abstract":"In spite of remarkable progress in deep latent variable generative modeling,\ntraining still remains a challenge due to a combination of optimization and\ngeneralization issues. In practice, a combination of heuristic algorithms (such\nas hand-crafted annealing of KL-terms) is often used in order to achieve the\ndesired results, but such solutions are not robust to changes in model\narchitecture or dataset. The best settings can often vary dramatically from one\nproblem to another, which requires doing expensive parameter sweeps for each\nnew case. Here we develop on the idea of training VAEs with additional\nconstraints as a way to control their behaviour. We first present a detailed\ntheoretical analysis of constrained VAEs, expanding our understanding of how\nthese models work. We then introduce and analyze a practical algorithm termed\nGeneralized ELBO with Constrained Optimization, GECO. The main advantage of\nGECO for the machine learning practitioner is a more intuitive, yet principled,\nprocess of tuning the loss. This involves defining of a set of constraints,\nwhich typically have an explicit relation to the desired model performance, in\ncontrast to tweaking abstract hyper-parameters which implicitly affect the\nmodel behavior. Encouraging experimental results in several standard datasets\nindicate that GECO is a very robust and effective tool to balance\nreconstruction and compression constraints.","url_abs":"http://arxiv.org/abs/1810.00597v1","url_pdf":"http://arxiv.org/pdf/1810.00597v1.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":"taming-vaes","repo_url":"https://github.com/denproc/Taming-VAEs","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"taming-vaes","repo_url":"https://github.com/pemami4911/EfficientMORL","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"taming-vaes","repo_url":"https://github.com/deepmind/sonnet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[],"methods":[{"method_slug":"geco","method_name":"GECO"}],"datasets_introduced":[],"methods_introduced":[{"slug":"geco","name":"GECO","full_name":"Generalized ELBO with Constrained Optimization"}],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1810.00597","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1810.00597"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/pemami4911/EfficientMORL","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/deepmind/sonnet","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/denproc/Taming-VAEs","reach":null}],"summary":{"ran_draft_wrong":3},"by_repo_kind":{"listed":{"samples":3,"ran":3,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":3,"samples":[{"code_sha256_prefix":"5fb1d231ba9d1ff8","entry":"IWAE_KL","repo":"denproc/Taming-VAEs","repo_kind":"listed","path":"IWAE_GECO.py","file_url":"https://github.com/denproc/Taming-VAEs/blob/HEAD/IWAE_GECO.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"5fb1d231ba9d1ff8"}},{"code_sha256_prefix":"c53f20354600989c","entry":"KL_divergence","repo":"denproc/Taming-VAEs","repo_kind":"listed","path":"IWAE_GECO.py","file_url":"https://github.com/denproc/Taming-VAEs/blob/HEAD/IWAE_GECO.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c53f20354600989c"}},{"code_sha256_prefix":"55afc47bb91691dd","entry":"RE","repo":"denproc/Taming-VAEs","repo_kind":"listed","path":"IWAE_GECO.py","file_url":"https://github.com/denproc/Taming-VAEs/blob/HEAD/IWAE_GECO.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"55afc47bb91691dd"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}