{"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/deep-ensembles-work-but-are-they-necessary","title":"Deep Ensembles Work, But Are They Necessary?","arxiv_id":"2202.06985","date":"2022-02-14","proceeding":null,"authors":["Taiga Abe","E. Kelly Buchanan","Geoff Pleiss","Richard Zemel","John P. Cunningham"],"abstract":"Ensembling neural networks is an effective way to increase accuracy, and can often match the performance of individual larger models. This observation poses a natural question: given the choice between a deep ensemble and a single neural network with similar accuracy, is one preferable over the other? Recent work suggests that deep ensembles may offer distinct benefits beyond predictive power: namely, uncertainty quantification and robustness to dataset shift. In this work, we demonstrate limitations to these purported benefits, and show that a single (but larger) neural network can replicate these qualities. First, we show that ensemble diversity, by any metric, does not meaningfully contribute to an ensemble's uncertainty quantification on out-of-distribution (OOD) data, but is instead highly correlated with the relative improvement of a single larger model. Second, we show that the OOD performance afforded by ensembles is strongly determined by their in-distribution (InD) performance, and -- in this sense -- is not indicative of any \"effective robustness\". While deep ensembles are a practical way to achieve improvements to predictive power, uncertainty quantification, and robustness, our results show that these improvements can be replicated by a (larger) single model.","url_abs":"https://arxiv.org/abs/2202.06985v2","url_pdf":"https://arxiv.org/pdf/2202.06985v2.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":"deep-ensembles-work-but-are-they-necessary","repo_url":"https://github.com/cellistigs/interp_ensembles","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"diversity","task_name":"Diversity"},{"task_slug":"uncertainty-quantification","task_name":"Uncertainty Quantification"}],"methods":[{"method_slug":"deep-ensembles","method_name":"Deep Ensembles"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2202.06985","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2202.06985"}},"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/cellistigs/interp_ensembles","reach":null}],"summary":{"ran_violates":3,"ran_fixture":1,"unverified":2},"by_repo_kind":{"official":{"samples":6,"ran":4,"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":6,"samples":[{"code_sha256_prefix":"ad05753c21667aac","entry":"JS","repo":"cellistigs/interp_ensembles","repo_kind":"official","path":"scripts/plot_conditional_variance.py","file_url":"https://github.com/cellistigs/interp_ensembles/blob/HEAD/scripts/plot_conditional_variance.py","link_basis":"first_harvest_node","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"ad05753c21667aac"}},{"code_sha256_prefix":"c8a654b055362cde","entry":"KL","repo":"cellistigs/interp_ensembles","repo_kind":"official","path":"scripts/plot_conditional_variance.py","file_url":"https://github.com/cellistigs/interp_ensembles/blob/HEAD/scripts/plot_conditional_variance.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c8a654b055362cde"}},{"code_sha256_prefix":"58b9d9f273880e5a","entry":"Var","repo":"cellistigs/interp_ensembles","repo_kind":"official","path":"scripts/plot_conditional_variance.py","file_url":"https://github.com/cellistigs/interp_ensembles/blob/HEAD/scripts/plot_conditional_variance.py","link_basis":"first_harvest_node","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"58b9d9f273880e5a"}},{"code_sha256_prefix":"2c8da032864c10a7","entry":"get_ordered_metricvals","repo":"cellistigs/interp_ensembles","repo_kind":"official","path":"scripts/paperfigs/compare_performance.py","file_url":"https://github.com/cellistigs/interp_ensembles/blob/HEAD/scripts/paperfigs/compare_performance.py","link_basis":"first_harvest_node","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"2c8da032864c10a7"}},{"code_sha256_prefix":"1c7277383b7833fc","entry":"get_ensemble","repo":"cellistigs/interp_ensembles","repo_kind":"official","path":"scripts/paperfigs/compare_performance.py","file_url":"https://github.com/cellistigs/interp_ensembles/blob/HEAD/scripts/paperfigs/compare_performance.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"1c7277383b7833fc"}},{"code_sha256_prefix":"4e4e868e1c440da2","entry":"get_metrics_outputs","repo":"cellistigs/interp_ensembles","repo_kind":"official","path":"scripts/paperfigs/compare_performance.py","file_url":"https://github.com/cellistigs/interp_ensembles/blob/HEAD/scripts/paperfigs/compare_performance.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"4e4e868e1c440da2"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}