{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:R7SEYTBZAWB5KOGULYWFHQH5RS","short_pith_number":"pith:R7SEYTBZ","schema_version":"1.0","canonical_sha256":"8fe44c4c390583d538d45e2c53c0fd8ca5d954929c92c24053917fba44358204","source":{"kind":"arxiv","id":"2202.12297","version":1},"attestation_state":"computed","paper":{"title":"Embedded Ensembles: Infinite Width Limit and Operating Regimes","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Alexey Zaytsev, Dmitry Yarotsky, Ivan Anokhin, Maksim Velikanov, Maxim Panov, Roman Kail, Roman Vashurin","submitted_at":"2022-02-24T18:55:41Z","abstract_excerpt":"A memory efficient approach to ensembling neural networks is to share most weights among the ensembled models by means of a single reference network. We refer to this strategy as Embedded Ensembling (EE); its particular examples are BatchEnsembles and Monte-Carlo dropout ensembles. In this paper we perform a systematic theoretical and empirical analysis of embedded ensembles with different number of models. Theoretically, we use a Neural-Tangent-Kernel-based approach to derive the wide network limit of the gradient descent dynamics. In this limit, we identify two ensemble regimes - independent"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2202.12297","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2022-02-24T18:55:41Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"321a6e4eec655bf8e472c06283848b9d256548815b04a14436b72f6ff9e3783f","abstract_canon_sha256":"82c987e99a38819b098331285e7805da1f3fd531a80eb66f5a9648d93be24ed4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:59:51.673560Z","signature_b64":"UsJeuXMAbLbC7nNt7FAlob2iLZt/LVhdhUijbG9tOwq35mIPNHCOcN5iW0GUnkR+xSZ+I5ZfwnJSChIt6qLzBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8fe44c4c390583d538d45e2c53c0fd8ca5d954929c92c24053917fba44358204","last_reissued_at":"2026-07-05T03:59:51.673116Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:59:51.673116Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Embedded Ensembles: Infinite Width Limit and Operating Regimes","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Alexey Zaytsev, Dmitry Yarotsky, Ivan Anokhin, Maksim Velikanov, Maxim Panov, Roman Kail, Roman Vashurin","submitted_at":"2022-02-24T18:55:41Z","abstract_excerpt":"A memory efficient approach to ensembling neural networks is to share most weights among the ensembled models by means of a single reference network. We refer to this strategy as Embedded Ensembling (EE); its particular examples are BatchEnsembles and Monte-Carlo dropout ensembles. In this paper we perform a systematic theoretical and empirical analysis of embedded ensembles with different number of models. Theoretically, we use a Neural-Tangent-Kernel-based approach to derive the wide network limit of the gradient descent dynamics. In this limit, we identify two ensemble regimes - independent"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2202.12297","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2202.12297/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2202.12297","created_at":"2026-07-05T03:59:51.673171+00:00"},{"alias_kind":"arxiv_version","alias_value":"2202.12297v1","created_at":"2026-07-05T03:59:51.673171+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2202.12297","created_at":"2026-07-05T03:59:51.673171+00:00"},{"alias_kind":"pith_short_12","alias_value":"R7SEYTBZAWB5","created_at":"2026-07-05T03:59:51.673171+00:00"},{"alias_kind":"pith_short_16","alias_value":"R7SEYTBZAWB5KOGU","created_at":"2026-07-05T03:59:51.673171+00:00"},{"alias_kind":"pith_short_8","alias_value":"R7SEYTBZ","created_at":"2026-07-05T03:59:51.673171+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/R7SEYTBZAWB5KOGULYWFHQH5RS","json":"https://pith.science/pith/R7SEYTBZAWB5KOGULYWFHQH5RS.json","graph_json":"https://pith.science/api/pith-number/R7SEYTBZAWB5KOGULYWFHQH5RS/graph.json","events_json":"https://pith.science/api/pith-number/R7SEYTBZAWB5KOGULYWFHQH5RS/events.json","paper":"https://pith.science/paper/R7SEYTBZ"},"agent_actions":{"view_html":"https://pith.science/pith/R7SEYTBZAWB5KOGULYWFHQH5RS","download_json":"https://pith.science/pith/R7SEYTBZAWB5KOGULYWFHQH5RS.json","view_paper":"https://pith.science/paper/R7SEYTBZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2202.12297&json=true","fetch_graph":"https://pith.science/api/pith-number/R7SEYTBZAWB5KOGULYWFHQH5RS/graph.json","fetch_events":"https://pith.science/api/pith-number/R7SEYTBZAWB5KOGULYWFHQH5RS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/R7SEYTBZAWB5KOGULYWFHQH5RS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/R7SEYTBZAWB5KOGULYWFHQH5RS/action/storage_attestation","attest_author":"https://pith.science/pith/R7SEYTBZAWB5KOGULYWFHQH5RS/action/author_attestation","sign_citation":"https://pith.science/pith/R7SEYTBZAWB5KOGULYWFHQH5RS/action/citation_signature","submit_replication":"https://pith.science/pith/R7SEYTBZAWB5KOGULYWFHQH5RS/action/replication_record"}},"created_at":"2026-07-05T03:59:51.673171+00:00","updated_at":"2026-07-05T03:59:51.673171+00:00"}