{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:ZLTGQZ2PHAKPSSJAJBHUWUV33X","short_pith_number":"pith:ZLTGQZ2P","schema_version":"1.0","canonical_sha256":"cae668674f3814f94920484f4b52bbdde733fd44e1c1b7369f9a41b175ddd302","source":{"kind":"arxiv","id":"1810.04650","version":2},"attestation_state":"computed","paper":{"title":"Multi-Task Learning as Multi-Objective Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Ozan Sener, Vladlen Koltun","submitted_at":"2018-10-10T17:18:09Z","abstract_excerpt":"In multi-task learning, multiple tasks are solved jointly, sharing inductive bias between them. Multi-task learning is inherently a multi-objective problem because different tasks may conflict, necessitating a trade-off. A common compromise is to optimize a proxy objective that minimizes a weighted linear combination of per-task losses. However, this workaround is only valid when the tasks do not compete, which is rarely the case. In this paper, we explicitly cast multi-task learning as multi-objective optimization, with the overall objective of finding a Pareto optimal solution. To this end, "},"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":"1810.04650","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-10-10T17:18:09Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"fff94599faf7b43fe283b261600e2d5411789c9566b5f54cac28444b570be27c","abstract_canon_sha256":"2dac7c4b6604083a32ff80f63a81a254f6fecc08399a4162d538b11c9a7c1c32"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:56:33.452933Z","signature_b64":"twrIioWv4w7lS45KdJDk7hgLjv5WWF7CiocCV5d6wm3IXXuPhMbRwZCSUqlsEu2nY+xLAee4PIkdd61JMMaXDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cae668674f3814f94920484f4b52bbdde733fd44e1c1b7369f9a41b175ddd302","last_reissued_at":"2026-05-17T23:56:33.451945Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:56:33.451945Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multi-Task Learning as Multi-Objective Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Ozan Sener, Vladlen Koltun","submitted_at":"2018-10-10T17:18:09Z","abstract_excerpt":"In multi-task learning, multiple tasks are solved jointly, sharing inductive bias between them. Multi-task learning is inherently a multi-objective problem because different tasks may conflict, necessitating a trade-off. A common compromise is to optimize a proxy objective that minimizes a weighted linear combination of per-task losses. However, this workaround is only valid when the tasks do not compete, which is rarely the case. In this paper, we explicitly cast multi-task learning as multi-objective optimization, with the overall objective of finding a Pareto optimal solution. To this end, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1810.04650","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"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":"1810.04650","created_at":"2026-05-17T23:56:33.452150+00:00"},{"alias_kind":"arxiv_version","alias_value":"1810.04650v2","created_at":"2026-05-17T23:56:33.452150+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1810.04650","created_at":"2026-05-17T23:56:33.452150+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZLTGQZ2PHAKP","created_at":"2026-05-18T12:33:07.085635+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZLTGQZ2PHAKPSSJA","created_at":"2026-05-18T12:33:07.085635+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZLTGQZ2P","created_at":"2026-05-18T12:33:07.085635+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2509.17189","citing_title":"Toward a unified data-driven turbulence model through multi-objective learning","ref_index":16,"is_internal_anchor":true},{"citing_arxiv_id":"2605.11654","citing_title":"Weather-Robust Cross-View Geo-Localization via Prototype-Based Semantic Part Discovery","ref_index":45,"is_internal_anchor":true},{"citing_arxiv_id":"2605.11654","citing_title":"Weather-Robust Cross-View Geo-Localization via Prototype-Based Semantic Part Discovery","ref_index":45,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZLTGQZ2PHAKPSSJAJBHUWUV33X","json":"https://pith.science/pith/ZLTGQZ2PHAKPSSJAJBHUWUV33X.json","graph_json":"https://pith.science/api/pith-number/ZLTGQZ2PHAKPSSJAJBHUWUV33X/graph.json","events_json":"https://pith.science/api/pith-number/ZLTGQZ2PHAKPSSJAJBHUWUV33X/events.json","paper":"https://pith.science/paper/ZLTGQZ2P"},"agent_actions":{"view_html":"https://pith.science/pith/ZLTGQZ2PHAKPSSJAJBHUWUV33X","download_json":"https://pith.science/pith/ZLTGQZ2PHAKPSSJAJBHUWUV33X.json","view_paper":"https://pith.science/paper/ZLTGQZ2P","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1810.04650&json=true","fetch_graph":"https://pith.science/api/pith-number/ZLTGQZ2PHAKPSSJAJBHUWUV33X/graph.json","fetch_events":"https://pith.science/api/pith-number/ZLTGQZ2PHAKPSSJAJBHUWUV33X/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZLTGQZ2PHAKPSSJAJBHUWUV33X/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZLTGQZ2PHAKPSSJAJBHUWUV33X/action/storage_attestation","attest_author":"https://pith.science/pith/ZLTGQZ2PHAKPSSJAJBHUWUV33X/action/author_attestation","sign_citation":"https://pith.science/pith/ZLTGQZ2PHAKPSSJAJBHUWUV33X/action/citation_signature","submit_replication":"https://pith.science/pith/ZLTGQZ2PHAKPSSJAJBHUWUV33X/action/replication_record"}},"created_at":"2026-05-17T23:56:33.452150+00:00","updated_at":"2026-05-17T23:56:33.452150+00:00"}