{"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/tighter-variational-bounds-are-not","title":"Tighter Variational Bounds are Not Necessarily Better","arxiv_id":"1802.04537","date":"2018-02-13","proceeding":"ICML 2018 7","authors":["Tom Rainforth","Adam R. Kosiorek","Tuan Anh Le","Chris J. Maddison","Maximilian Igl","Frank Wood","Yee Whye Teh"],"abstract":"We provide theoretical and empirical evidence that using tighter evidence\nlower bounds (ELBOs) can be detrimental to the process of learning an inference\nnetwork by reducing the signal-to-noise ratio of the gradient estimator. Our\nresults call into question common implicit assumptions that tighter ELBOs are\nbetter variational objectives for simultaneous model learning and inference\namortization schemes. Based on our insights, we introduce three new algorithms:\nthe partially importance weighted auto-encoder (PIWAE), the multiply importance\nweighted auto-encoder (MIWAE), and the combination importance weighted\nauto-encoder (CIWAE), each of which includes the standard importance weighted\nauto-encoder (IWAE) as a special case. We show that each can deliver\nimprovements over IWAE, even when performance is measured by the IWAE target\nitself. Furthermore, our results suggest that PIWAE may be able to deliver\nsimultaneous improvements in the training of both the inference and generative\nnetworks.","url_abs":"http://arxiv.org/abs/1802.04537v3","url_pdf":"http://arxiv.org/pdf/1802.04537v3.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":"tighter-variational-bounds-are-not","repo_url":"https://github.com/CharlesArnal/IWAE_replication_project","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"tighter-variational-bounds-are-not","repo_url":"https://github.com/clementchadebec/benchmark_VAE","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"tighter-variational-bounds-are-not","repo_url":"https://github.com/madhubabuv/TightIWAE","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1802.04537","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1802.04537"}},"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/clementchadebec/benchmark_VAE","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/CharlesArnal/IWAE_replication_project","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/madhubabuv/TightIWAE","reach":null}],"summary":{"ran_draft_wrong":1},"by_repo_kind":{"listed":{"samples":1,"ran":1,"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":1,"samples":[{"code_sha256_prefix":"edbb10d386cd841b","entry":"debug_shape","repo":"madhubabuv/TightIWAE","repo_kind":"listed","path":"miwae_simplified.py","file_url":"https://github.com/madhubabuv/TightIWAE/blob/HEAD/miwae_simplified.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":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"edbb10d386cd841b"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}