{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:WJDNYC7FHRYAJTDRXBKXQP2FZT","short_pith_number":"pith:WJDNYC7F","schema_version":"1.0","canonical_sha256":"b246dc0be53c7004cc71b855783f45ccf878c1a27c3e1523d8b93a3056623190","source":{"kind":"arxiv","id":"2212.11268","version":1},"attestation_state":"computed","paper":{"title":"Personalized Decentralized Multi-Task Learning Over Dynamic Communication Graphs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.IT","eess.SP","math.IT"],"primary_cat":"cs.LG","authors_text":"Matin Mortaheb, Sennur Ulukus","submitted_at":"2022-12-21T18:58:24Z","abstract_excerpt":"Decentralized and federated learning algorithms face data heterogeneity as one of the biggest challenges, especially when users want to learn a specific task. Even when personalized headers are used concatenated to a shared network (PF-MTL), aggregating all the networks with a decentralized algorithm can result in performance degradation as a result of heterogeneity in the data. Our algorithm uses exchanged gradients to calculate the correlations among tasks automatically, and dynamically adjusts the communication graph to connect mutually beneficial tasks and isolate those that may negatively"},"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":"2212.11268","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-12-21T18:58:24Z","cross_cats_sorted":["cs.IT","eess.SP","math.IT"],"title_canon_sha256":"ccbbd76f05e47178dd6f7965391636a9c151816cd1fad3e8f7ed5c555bcb8212","abstract_canon_sha256":"e824113ee4d17c06f0320014f08d6db42e9b6729a42200cbee21bb64e98b0bc5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:27:28.954565Z","signature_b64":"t4X94YW/xYGNeAhDWsZ5wsgb6geGjts92pDI9ydHsrmk3uMkewHEOHUb2ZX0dBvAwHHC76ygDQPuDdDiqkxIAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b246dc0be53c7004cc71b855783f45ccf878c1a27c3e1523d8b93a3056623190","last_reissued_at":"2026-07-05T05:27:28.954002Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:27:28.954002Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Personalized Decentralized Multi-Task Learning Over Dynamic Communication Graphs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.IT","eess.SP","math.IT"],"primary_cat":"cs.LG","authors_text":"Matin Mortaheb, Sennur Ulukus","submitted_at":"2022-12-21T18:58:24Z","abstract_excerpt":"Decentralized and federated learning algorithms face data heterogeneity as one of the biggest challenges, especially when users want to learn a specific task. Even when personalized headers are used concatenated to a shared network (PF-MTL), aggregating all the networks with a decentralized algorithm can result in performance degradation as a result of heterogeneity in the data. Our algorithm uses exchanged gradients to calculate the correlations among tasks automatically, and dynamically adjusts the communication graph to connect mutually beneficial tasks and isolate those that may negatively"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2212.11268","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/2212.11268/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":"2212.11268","created_at":"2026-07-05T05:27:28.954059+00:00"},{"alias_kind":"arxiv_version","alias_value":"2212.11268v1","created_at":"2026-07-05T05:27:28.954059+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2212.11268","created_at":"2026-07-05T05:27:28.954059+00:00"},{"alias_kind":"pith_short_12","alias_value":"WJDNYC7FHRYA","created_at":"2026-07-05T05:27:28.954059+00:00"},{"alias_kind":"pith_short_16","alias_value":"WJDNYC7FHRYAJTDR","created_at":"2026-07-05T05:27:28.954059+00:00"},{"alias_kind":"pith_short_8","alias_value":"WJDNYC7F","created_at":"2026-07-05T05:27:28.954059+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/WJDNYC7FHRYAJTDRXBKXQP2FZT","json":"https://pith.science/pith/WJDNYC7FHRYAJTDRXBKXQP2FZT.json","graph_json":"https://pith.science/api/pith-number/WJDNYC7FHRYAJTDRXBKXQP2FZT/graph.json","events_json":"https://pith.science/api/pith-number/WJDNYC7FHRYAJTDRXBKXQP2FZT/events.json","paper":"https://pith.science/paper/WJDNYC7F"},"agent_actions":{"view_html":"https://pith.science/pith/WJDNYC7FHRYAJTDRXBKXQP2FZT","download_json":"https://pith.science/pith/WJDNYC7FHRYAJTDRXBKXQP2FZT.json","view_paper":"https://pith.science/paper/WJDNYC7F","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2212.11268&json=true","fetch_graph":"https://pith.science/api/pith-number/WJDNYC7FHRYAJTDRXBKXQP2FZT/graph.json","fetch_events":"https://pith.science/api/pith-number/WJDNYC7FHRYAJTDRXBKXQP2FZT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WJDNYC7FHRYAJTDRXBKXQP2FZT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WJDNYC7FHRYAJTDRXBKXQP2FZT/action/storage_attestation","attest_author":"https://pith.science/pith/WJDNYC7FHRYAJTDRXBKXQP2FZT/action/author_attestation","sign_citation":"https://pith.science/pith/WJDNYC7FHRYAJTDRXBKXQP2FZT/action/citation_signature","submit_replication":"https://pith.science/pith/WJDNYC7FHRYAJTDRXBKXQP2FZT/action/replication_record"}},"created_at":"2026-07-05T05:27:28.954059+00:00","updated_at":"2026-07-05T05:27:28.954059+00:00"}