{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:JFHO755Y5EQSAW5RRMG42VDXFN","short_pith_number":"pith:JFHO755Y","schema_version":"1.0","canonical_sha256":"494eeff7b8e921205bb18b0dcd54772b6c1220c6a91a2defa0570013033dbaec","source":{"kind":"arxiv","id":"2010.13723","version":3},"attestation_state":"computed","paper":{"title":"Optimal Client Sampling for Federated Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.LG","authors_text":"Peter Richtarik, Samuel Horvath, Wenlin Chen","submitted_at":"2020-10-26T17:05:13Z","abstract_excerpt":"It is well understood that client-master communication can be a primary bottleneck in Federated Learning. In this work, we address this issue with a novel client subsampling scheme, where we restrict the number of clients allowed to communicate their updates back to the master node. In each communication round, all participating clients compute their updates, but only the ones with \"important\" updates communicate back to the master. We show that importance can be measured using only the norm of the update and give a formula for optimal client participation. This formula minimizes the distance "},"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":"2010.13723","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-10-26T17:05:13Z","cross_cats_sorted":["cs.DC"],"title_canon_sha256":"68709c3c03e679419d677250375227e1860667dcfaf382b4ee3d19f16c92f27e","abstract_canon_sha256":"0e8a2ad9c53f580e6c0877c5ee2f188e53dffa121a897a059522b0e481f86284"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:50:09.509840Z","signature_b64":"AvYtk8yklhHwnhJ8FTu3juk/SrHci/kXlhsUNstQjtiBrIl4+lzeF146AtfQ9own2JJsr1Is8MTq2rR868YKBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"494eeff7b8e921205bb18b0dcd54772b6c1220c6a91a2defa0570013033dbaec","last_reissued_at":"2026-07-05T04:50:09.509483Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:50:09.509483Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Optimal Client Sampling for Federated Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.LG","authors_text":"Peter Richtarik, Samuel Horvath, Wenlin Chen","submitted_at":"2020-10-26T17:05:13Z","abstract_excerpt":"It is well understood that client-master communication can be a primary bottleneck in Federated Learning. In this work, we address this issue with a novel client subsampling scheme, where we restrict the number of clients allowed to communicate their updates back to the master node. In each communication round, all participating clients compute their updates, but only the ones with \"important\" updates communicate back to the master. We show that importance can be measured using only the norm of the update and give a formula for optimal client participation. This formula minimizes the distance "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.13723","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/2010.13723/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":"2010.13723","created_at":"2026-07-05T04:50:09.509541+00:00"},{"alias_kind":"arxiv_version","alias_value":"2010.13723v3","created_at":"2026-07-05T04:50:09.509541+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.13723","created_at":"2026-07-05T04:50:09.509541+00:00"},{"alias_kind":"pith_short_12","alias_value":"JFHO755Y5EQS","created_at":"2026-07-05T04:50:09.509541+00:00"},{"alias_kind":"pith_short_16","alias_value":"JFHO755Y5EQSAW5R","created_at":"2026-07-05T04:50:09.509541+00:00"},{"alias_kind":"pith_short_8","alias_value":"JFHO755Y","created_at":"2026-07-05T04:50:09.509541+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.10124","citing_title":"FedSteer: Taming Extreme Gradient Staleness in Federated Learning with Corrective Projections and Caching","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2509.17398","citing_title":"Optimizing Split Federated Learning with Unstable Client Participation","ref_index":33,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JFHO755Y5EQSAW5RRMG42VDXFN","json":"https://pith.science/pith/JFHO755Y5EQSAW5RRMG42VDXFN.json","graph_json":"https://pith.science/api/pith-number/JFHO755Y5EQSAW5RRMG42VDXFN/graph.json","events_json":"https://pith.science/api/pith-number/JFHO755Y5EQSAW5RRMG42VDXFN/events.json","paper":"https://pith.science/paper/JFHO755Y"},"agent_actions":{"view_html":"https://pith.science/pith/JFHO755Y5EQSAW5RRMG42VDXFN","download_json":"https://pith.science/pith/JFHO755Y5EQSAW5RRMG42VDXFN.json","view_paper":"https://pith.science/paper/JFHO755Y","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2010.13723&json=true","fetch_graph":"https://pith.science/api/pith-number/JFHO755Y5EQSAW5RRMG42VDXFN/graph.json","fetch_events":"https://pith.science/api/pith-number/JFHO755Y5EQSAW5RRMG42VDXFN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JFHO755Y5EQSAW5RRMG42VDXFN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JFHO755Y5EQSAW5RRMG42VDXFN/action/storage_attestation","attest_author":"https://pith.science/pith/JFHO755Y5EQSAW5RRMG42VDXFN/action/author_attestation","sign_citation":"https://pith.science/pith/JFHO755Y5EQSAW5RRMG42VDXFN/action/citation_signature","submit_replication":"https://pith.science/pith/JFHO755Y5EQSAW5RRMG42VDXFN/action/replication_record"}},"created_at":"2026-07-05T04:50:09.509541+00:00","updated_at":"2026-07-05T04:50:09.509541+00:00"}