{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:Y3JFSUIPLAF4HCLUZOQUTIYBGV","short_pith_number":"pith:Y3JFSUIP","schema_version":"1.0","canonical_sha256":"c6d259510f580bc38974cba149a3013578ad763ce971dfd9fb9b4bcabc24af24","source":{"kind":"arxiv","id":"2010.03622","version":5},"attestation_state":"computed","paper":{"title":"Theoretical Analysis of Self-Training with Deep Networks on Unlabeled Data","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Colin Wei, Kendrick Shen, Tengyu Ma, Yining Chen","submitted_at":"2020-10-07T19:43:55Z","abstract_excerpt":"Self-training algorithms, which train a model to fit pseudolabels predicted by another previously-learned model, have been very successful for learning with unlabeled data using neural networks. However, the current theoretical understanding of self-training only applies to linear models. This work provides a unified theoretical analysis of self-training with deep networks for semi-supervised learning, unsupervised domain adaptation, and unsupervised learning. At the core of our analysis is a simple but realistic \"expansion\" assumption, which states that a low probability subset of the data mu"},"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":"2010.03622","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-10-07T19:43:55Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"998f6f0dcf89b9875aa71fcfa659afbedc53d480dc147e4f765d14fe86eefe52","abstract_canon_sha256":"b8538966c80b6856b4cd2b3c1fd6765653602b003922240ecc5171d580cb7e09"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:16:34.560641Z","signature_b64":"ZmIdZWR4Rv4Nx8k4/GmC1QS6oyYxonZ+1YTcMR/DsZ4YE8JQWxh6ukixF8DzjTdCoWjmVHMtmsE20w6CwwVqBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c6d259510f580bc38974cba149a3013578ad763ce971dfd9fb9b4bcabc24af24","last_reissued_at":"2026-07-05T04:16:34.560153Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:16:34.560153Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Theoretical Analysis of Self-Training with Deep Networks on Unlabeled Data","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Colin Wei, Kendrick Shen, Tengyu Ma, Yining Chen","submitted_at":"2020-10-07T19:43:55Z","abstract_excerpt":"Self-training algorithms, which train a model to fit pseudolabels predicted by another previously-learned model, have been very successful for learning with unlabeled data using neural networks. However, the current theoretical understanding of self-training only applies to linear models. This work provides a unified theoretical analysis of self-training with deep networks for semi-supervised learning, unsupervised domain adaptation, and unsupervised learning. At the core of our analysis is a simple but realistic \"expansion\" assumption, which states that a low probability subset of the data mu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.03622","kind":"arxiv","version":5},"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/2010.03622/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":"2010.03622","created_at":"2026-07-05T04:16:34.560212+00:00"},{"alias_kind":"arxiv_version","alias_value":"2010.03622v5","created_at":"2026-07-05T04:16:34.560212+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.03622","created_at":"2026-07-05T04:16:34.560212+00:00"},{"alias_kind":"pith_short_12","alias_value":"Y3JFSUIPLAF4","created_at":"2026-07-05T04:16:34.560212+00:00"},{"alias_kind":"pith_short_16","alias_value":"Y3JFSUIPLAF4HCLU","created_at":"2026-07-05T04:16:34.560212+00:00"},{"alias_kind":"pith_short_8","alias_value":"Y3JFSUIP","created_at":"2026-07-05T04:16:34.560212+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2502.05075","citing_title":"Discrepancies are Virtue: Weak-to-Strong Generalization through Lens of Intrinsic Dimension","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17778","citing_title":"Self-Distillation is Optimal Among Spectral Shrinkage Estimators in Spiked Covariance Models","ref_index":65,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11857","citing_title":"Beyond Parameter Aggregation: Semantic Consensus for Federated Fine-Tuning of LLMs","ref_index":32,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Y3JFSUIPLAF4HCLUZOQUTIYBGV","json":"https://pith.science/pith/Y3JFSUIPLAF4HCLUZOQUTIYBGV.json","graph_json":"https://pith.science/api/pith-number/Y3JFSUIPLAF4HCLUZOQUTIYBGV/graph.json","events_json":"https://pith.science/api/pith-number/Y3JFSUIPLAF4HCLUZOQUTIYBGV/events.json","paper":"https://pith.science/paper/Y3JFSUIP"},"agent_actions":{"view_html":"https://pith.science/pith/Y3JFSUIPLAF4HCLUZOQUTIYBGV","download_json":"https://pith.science/pith/Y3JFSUIPLAF4HCLUZOQUTIYBGV.json","view_paper":"https://pith.science/paper/Y3JFSUIP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2010.03622&json=true","fetch_graph":"https://pith.science/api/pith-number/Y3JFSUIPLAF4HCLUZOQUTIYBGV/graph.json","fetch_events":"https://pith.science/api/pith-number/Y3JFSUIPLAF4HCLUZOQUTIYBGV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Y3JFSUIPLAF4HCLUZOQUTIYBGV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Y3JFSUIPLAF4HCLUZOQUTIYBGV/action/storage_attestation","attest_author":"https://pith.science/pith/Y3JFSUIPLAF4HCLUZOQUTIYBGV/action/author_attestation","sign_citation":"https://pith.science/pith/Y3JFSUIPLAF4HCLUZOQUTIYBGV/action/citation_signature","submit_replication":"https://pith.science/pith/Y3JFSUIPLAF4HCLUZOQUTIYBGV/action/replication_record"}},"created_at":"2026-07-05T04:16:34.560212+00:00","updated_at":"2026-07-05T04:16:34.560212+00:00"}