{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:RDDLS6PM5N5DUXVFXXMXH5PSX7","short_pith_number":"pith:RDDLS6PM","schema_version":"1.0","canonical_sha256":"88c6b979eceb7a3a5ea5bdd973f5f2bfd3e087fa90485bdc561cb8fb2fa4e2d8","source":{"kind":"arxiv","id":"2109.07704","version":4},"attestation_state":"computed","paper":{"title":"Federated Submodel Optimization for Hot and Cold Data Features","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Chaoyue Niu, Chengfei Lv, Fan Wu, Guihai Chen, Shaojie Tang, Yanghe Feng, Yucheng Ding","submitted_at":"2021-09-16T03:54:54Z","abstract_excerpt":"We study practical data characteristics underlying federated learning, where non-i.i.d. data from clients have sparse features, and a certain client's local data normally involves only a small part of the full model, called a submodel. Due to data sparsity, the classical federated averaging (FedAvg) algorithm or its variants will be severely slowed down, because when updating the global model, each client's zero update of the full model excluding its submodel is inaccurately aggregated. Therefore, we propose federated submodel averaging (FedSubAvg), ensuring that the expectation of the global "},"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":"2109.07704","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-09-16T03:54:54Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"5e743858a82aef22669865aecb6ef5eda2a901e430f493c1f3dfbd27f192b978","abstract_canon_sha256":"4cdfeadf760751aa2359d1df094c06af2bf61bb5a932e5cd312413f2bc27adc6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:58:14.644862Z","signature_b64":"5PZBrGBNVZW2153m1BlOLfrBb+tfFjysto2i/balJm1yTGEx6PFK7iF0c28gOAcc13OMHUHX5RTS2HB7ZjVNBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"88c6b979eceb7a3a5ea5bdd973f5f2bfd3e087fa90485bdc561cb8fb2fa4e2d8","last_reissued_at":"2026-07-05T05:58:14.644444Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:58:14.644444Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Federated Submodel Optimization for Hot and Cold Data Features","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Chaoyue Niu, Chengfei Lv, Fan Wu, Guihai Chen, Shaojie Tang, Yanghe Feng, Yucheng Ding","submitted_at":"2021-09-16T03:54:54Z","abstract_excerpt":"We study practical data characteristics underlying federated learning, where non-i.i.d. data from clients have sparse features, and a certain client's local data normally involves only a small part of the full model, called a submodel. Due to data sparsity, the classical federated averaging (FedAvg) algorithm or its variants will be severely slowed down, because when updating the global model, each client's zero update of the full model excluding its submodel is inaccurately aggregated. Therefore, we propose federated submodel averaging (FedSubAvg), ensuring that the expectation of the global "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2109.07704","kind":"arxiv","version":4},"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/2109.07704/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":"2109.07704","created_at":"2026-07-05T05:58:14.644503+00:00"},{"alias_kind":"arxiv_version","alias_value":"2109.07704v4","created_at":"2026-07-05T05:58:14.644503+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2109.07704","created_at":"2026-07-05T05:58:14.644503+00:00"},{"alias_kind":"pith_short_12","alias_value":"RDDLS6PM5N5D","created_at":"2026-07-05T05:58:14.644503+00:00"},{"alias_kind":"pith_short_16","alias_value":"RDDLS6PM5N5DUXVF","created_at":"2026-07-05T05:58:14.644503+00:00"},{"alias_kind":"pith_short_8","alias_value":"RDDLS6PM","created_at":"2026-07-05T05:58:14.644503+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/RDDLS6PM5N5DUXVFXXMXH5PSX7","json":"https://pith.science/pith/RDDLS6PM5N5DUXVFXXMXH5PSX7.json","graph_json":"https://pith.science/api/pith-number/RDDLS6PM5N5DUXVFXXMXH5PSX7/graph.json","events_json":"https://pith.science/api/pith-number/RDDLS6PM5N5DUXVFXXMXH5PSX7/events.json","paper":"https://pith.science/paper/RDDLS6PM"},"agent_actions":{"view_html":"https://pith.science/pith/RDDLS6PM5N5DUXVFXXMXH5PSX7","download_json":"https://pith.science/pith/RDDLS6PM5N5DUXVFXXMXH5PSX7.json","view_paper":"https://pith.science/paper/RDDLS6PM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2109.07704&json=true","fetch_graph":"https://pith.science/api/pith-number/RDDLS6PM5N5DUXVFXXMXH5PSX7/graph.json","fetch_events":"https://pith.science/api/pith-number/RDDLS6PM5N5DUXVFXXMXH5PSX7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RDDLS6PM5N5DUXVFXXMXH5PSX7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RDDLS6PM5N5DUXVFXXMXH5PSX7/action/storage_attestation","attest_author":"https://pith.science/pith/RDDLS6PM5N5DUXVFXXMXH5PSX7/action/author_attestation","sign_citation":"https://pith.science/pith/RDDLS6PM5N5DUXVFXXMXH5PSX7/action/citation_signature","submit_replication":"https://pith.science/pith/RDDLS6PM5N5DUXVFXXMXH5PSX7/action/replication_record"}},"created_at":"2026-07-05T05:58:14.644503+00:00","updated_at":"2026-07-05T05:58:14.644503+00:00"}