{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:U3FRGCMU3JYZY4LBBRZYMHH7SB","short_pith_number":"pith:U3FRGCMU","schema_version":"1.0","canonical_sha256":"a6cb130994da719c71610c73861cff9062fbe0ec906f0491dc231d93147ff744","source":{"kind":"arxiv","id":"2211.09981","version":3},"attestation_state":"computed","paper":{"title":"Weighted Ensemble Self-Supervised Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Alexander A. Alemi, Ian Fischer, Joshua V. Dillon, Saurabh Singh, Sergey Ioffe, Warren Morningstar, Yangjun Ruan","submitted_at":"2022-11-18T02:00:17Z","abstract_excerpt":"Ensembling has proven to be a powerful technique for boosting model performance, uncertainty estimation, and robustness in supervised learning. Advances in self-supervised learning (SSL) enable leveraging large unlabeled corpora for state-of-the-art few-shot and supervised learning performance. In this paper, we explore how ensemble methods can improve recent SSL techniques by developing a framework that permits data-dependent weighted cross-entropy losses. We refrain from ensembling the representation backbone; this choice yields an efficient ensemble method that incurs a small training cost "},"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":"2211.09981","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-11-18T02:00:17Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"985929461b809aa6255adc3e0e0a7253771481cb2d4fea420b3cafc0954e2105","abstract_canon_sha256":"12cffad4938d65420838a93caf9458c453d510ac1b7852e2ca6e1e1e1757a632"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:59:22.064114Z","signature_b64":"saCpQATRfGLKn9os9n//ZPqfIe6lbnqEWXnmt9f6LzOBy5NnJVwPoTnjH/Fg6AML3t9tO811gfs+TBU2/fNaCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a6cb130994da719c71610c73861cff9062fbe0ec906f0491dc231d93147ff744","last_reissued_at":"2026-07-05T05:59:22.063688Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:59:22.063688Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Weighted Ensemble Self-Supervised Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Alexander A. Alemi, Ian Fischer, Joshua V. Dillon, Saurabh Singh, Sergey Ioffe, Warren Morningstar, Yangjun Ruan","submitted_at":"2022-11-18T02:00:17Z","abstract_excerpt":"Ensembling has proven to be a powerful technique for boosting model performance, uncertainty estimation, and robustness in supervised learning. Advances in self-supervised learning (SSL) enable leveraging large unlabeled corpora for state-of-the-art few-shot and supervised learning performance. In this paper, we explore how ensemble methods can improve recent SSL techniques by developing a framework that permits data-dependent weighted cross-entropy losses. We refrain from ensembling the representation backbone; this choice yields an efficient ensemble method that incurs a small training cost "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2211.09981","kind":"arxiv","version":3},"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/2211.09981/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":"2211.09981","created_at":"2026-07-05T05:59:22.063742+00:00"},{"alias_kind":"arxiv_version","alias_value":"2211.09981v3","created_at":"2026-07-05T05:59:22.063742+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2211.09981","created_at":"2026-07-05T05:59:22.063742+00:00"},{"alias_kind":"pith_short_12","alias_value":"U3FRGCMU3JYZ","created_at":"2026-07-05T05:59:22.063742+00:00"},{"alias_kind":"pith_short_16","alias_value":"U3FRGCMU3JYZY4LB","created_at":"2026-07-05T05:59:22.063742+00:00"},{"alias_kind":"pith_short_8","alias_value":"U3FRGCMU","created_at":"2026-07-05T05:59:22.063742+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.01566","citing_title":"Multi-objective Deep Learning: Taxonomy and Survey of the State of the Art","ref_index":149,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/U3FRGCMU3JYZY4LBBRZYMHH7SB","json":"https://pith.science/pith/U3FRGCMU3JYZY4LBBRZYMHH7SB.json","graph_json":"https://pith.science/api/pith-number/U3FRGCMU3JYZY4LBBRZYMHH7SB/graph.json","events_json":"https://pith.science/api/pith-number/U3FRGCMU3JYZY4LBBRZYMHH7SB/events.json","paper":"https://pith.science/paper/U3FRGCMU"},"agent_actions":{"view_html":"https://pith.science/pith/U3FRGCMU3JYZY4LBBRZYMHH7SB","download_json":"https://pith.science/pith/U3FRGCMU3JYZY4LBBRZYMHH7SB.json","view_paper":"https://pith.science/paper/U3FRGCMU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2211.09981&json=true","fetch_graph":"https://pith.science/api/pith-number/U3FRGCMU3JYZY4LBBRZYMHH7SB/graph.json","fetch_events":"https://pith.science/api/pith-number/U3FRGCMU3JYZY4LBBRZYMHH7SB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/U3FRGCMU3JYZY4LBBRZYMHH7SB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/U3FRGCMU3JYZY4LBBRZYMHH7SB/action/storage_attestation","attest_author":"https://pith.science/pith/U3FRGCMU3JYZY4LBBRZYMHH7SB/action/author_attestation","sign_citation":"https://pith.science/pith/U3FRGCMU3JYZY4LBBRZYMHH7SB/action/citation_signature","submit_replication":"https://pith.science/pith/U3FRGCMU3JYZY4LBBRZYMHH7SB/action/replication_record"}},"created_at":"2026-07-05T05:59:22.063742+00:00","updated_at":"2026-07-05T05:59:22.063742+00:00"}