{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:KUQDNMFHZFAKO4TJZO3Z2WV6BD","short_pith_number":"pith:KUQDNMFH","schema_version":"1.0","canonical_sha256":"552036b0a7c940a77269cbb79d5abe08ff08c9f96cefcb8ba435a657a5646559","source":{"kind":"arxiv","id":"2408.07985","version":1},"attestation_state":"computed","paper":{"title":"Analytical Uncertainty-Based Loss Weighting in Multi-Task Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Cathrin Elich, Heiner Stuckenschmidt, Jan M. K\\\"ohler, Lukas Kirchdorfer, Lukas Schott, Simon Kutsche","submitted_at":"2024-08-15T07:10:17Z","abstract_excerpt":"With the rise of neural networks in various domains, multi-task learning (MTL) gained significant relevance. A key challenge in MTL is balancing individual task losses during neural network training to improve performance and efficiency through knowledge sharing across tasks. To address these challenges, we propose a novel task-weighting method by building on the most prevalent approach of Uncertainty Weighting and computing analytically optimal uncertainty-based weights, normalized by a softmax function with tunable temperature. Our approach yields comparable results to the combinatorially pr"},"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":"2408.07985","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-08-15T07:10:17Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"71baa56f33066c6ccef8c1993efc96ed63736b2781da1bfc61bda4052298644a","abstract_canon_sha256":"73c61c2a03c6dd526bd1ad5b3c7fb48a19a8b9901fd001dc8f9682cd03130dc1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:55:42.844412Z","signature_b64":"ATNNHEfQ6FVKuo0m4lPN9aaIZeog+HRbmS2PmdroOh0ImMZBvP2UiTy9zWD2R78mLj4iqDzBy4NTZ/Y9BFgRAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"552036b0a7c940a77269cbb79d5abe08ff08c9f96cefcb8ba435a657a5646559","last_reissued_at":"2026-07-05T08:55:42.843971Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:55:42.843971Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Analytical Uncertainty-Based Loss Weighting in Multi-Task Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Cathrin Elich, Heiner Stuckenschmidt, Jan M. K\\\"ohler, Lukas Kirchdorfer, Lukas Schott, Simon Kutsche","submitted_at":"2024-08-15T07:10:17Z","abstract_excerpt":"With the rise of neural networks in various domains, multi-task learning (MTL) gained significant relevance. A key challenge in MTL is balancing individual task losses during neural network training to improve performance and efficiency through knowledge sharing across tasks. To address these challenges, we propose a novel task-weighting method by building on the most prevalent approach of Uncertainty Weighting and computing analytically optimal uncertainty-based weights, normalized by a softmax function with tunable temperature. Our approach yields comparable results to the combinatorially pr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.07985","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/2408.07985/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":"2408.07985","created_at":"2026-07-05T08:55:42.844027+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.07985v1","created_at":"2026-07-05T08:55:42.844027+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.07985","created_at":"2026-07-05T08:55:42.844027+00:00"},{"alias_kind":"pith_short_12","alias_value":"KUQDNMFHZFAK","created_at":"2026-07-05T08:55:42.844027+00:00"},{"alias_kind":"pith_short_16","alias_value":"KUQDNMFHZFAKO4TJ","created_at":"2026-07-05T08:55:42.844027+00:00"},{"alias_kind":"pith_short_8","alias_value":"KUQDNMFH","created_at":"2026-07-05T08:55:42.844027+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2411.10077","citing_title":"Hierarchical mutual distillation for multi-view fusion: Learning from all possible view combinations","ref_index":33,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KUQDNMFHZFAKO4TJZO3Z2WV6BD","json":"https://pith.science/pith/KUQDNMFHZFAKO4TJZO3Z2WV6BD.json","graph_json":"https://pith.science/api/pith-number/KUQDNMFHZFAKO4TJZO3Z2WV6BD/graph.json","events_json":"https://pith.science/api/pith-number/KUQDNMFHZFAKO4TJZO3Z2WV6BD/events.json","paper":"https://pith.science/paper/KUQDNMFH"},"agent_actions":{"view_html":"https://pith.science/pith/KUQDNMFHZFAKO4TJZO3Z2WV6BD","download_json":"https://pith.science/pith/KUQDNMFHZFAKO4TJZO3Z2WV6BD.json","view_paper":"https://pith.science/paper/KUQDNMFH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.07985&json=true","fetch_graph":"https://pith.science/api/pith-number/KUQDNMFHZFAKO4TJZO3Z2WV6BD/graph.json","fetch_events":"https://pith.science/api/pith-number/KUQDNMFHZFAKO4TJZO3Z2WV6BD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KUQDNMFHZFAKO4TJZO3Z2WV6BD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KUQDNMFHZFAKO4TJZO3Z2WV6BD/action/storage_attestation","attest_author":"https://pith.science/pith/KUQDNMFHZFAKO4TJZO3Z2WV6BD/action/author_attestation","sign_citation":"https://pith.science/pith/KUQDNMFHZFAKO4TJZO3Z2WV6BD/action/citation_signature","submit_replication":"https://pith.science/pith/KUQDNMFHZFAKO4TJZO3Z2WV6BD/action/replication_record"}},"created_at":"2026-07-05T08:55:42.844027+00:00","updated_at":"2026-07-05T08:55:42.844027+00:00"}