{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:JZ567H7WDJKX6HNAMKUNCO67CQ","short_pith_number":"pith:JZ567H7W","schema_version":"1.0","canonical_sha256":"4e7bef9ff61a557f1da062a8d13bdf14003863f906755d6040b07865bf13776a","source":{"kind":"arxiv","id":"2007.02693","version":3},"attestation_state":"computed","paper":{"title":"Auxiliary Learning by Implicit Differentiation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cs.CV","authors_text":"Aviv Navon, Ethan Fetaya, Gal Chechik, Haggai Maron, Idan Achituve","submitted_at":"2020-06-22T19:35:07Z","abstract_excerpt":"Training neural networks with auxiliary tasks is a common practice for improving the performance on a main task of interest. Two main challenges arise in this multi-task learning setting: (i) designing useful auxiliary tasks; and (ii) combining auxiliary tasks into a single coherent loss. Here, we propose a novel framework, AuxiLearn, that targets both challenges based on implicit differentiation. First, when useful auxiliaries are known, we propose learning a network that combines all losses into a single coherent objective function. This network can learn non-linear interactions between task"},"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":"2007.02693","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2020-06-22T19:35:07Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"810644c8dc969e0a6fc65017631a85fbf08c7b350944c7c21901c98ffa7cb7ff","abstract_canon_sha256":"4d424d44e5d6655f6e2f6d5b3bef464c9f2e5457ced8de51198f6f2e2c2f659d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:39:02.752908Z","signature_b64":"XNg7jmJaD9mJTExBSFS86FKixKLkZjf6jUeQ0o8cqJbQekx5wevdSCh3mNKlt7gZRxtf0iDw8Bun9KkMtuCTAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4e7bef9ff61a557f1da062a8d13bdf14003863f906755d6040b07865bf13776a","last_reissued_at":"2026-07-05T02:39:02.752419Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:39:02.752419Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Auxiliary Learning by Implicit Differentiation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cs.CV","authors_text":"Aviv Navon, Ethan Fetaya, Gal Chechik, Haggai Maron, Idan Achituve","submitted_at":"2020-06-22T19:35:07Z","abstract_excerpt":"Training neural networks with auxiliary tasks is a common practice for improving the performance on a main task of interest. Two main challenges arise in this multi-task learning setting: (i) designing useful auxiliary tasks; and (ii) combining auxiliary tasks into a single coherent loss. Here, we propose a novel framework, AuxiLearn, that targets both challenges based on implicit differentiation. First, when useful auxiliaries are known, we propose learning a network that combines all losses into a single coherent objective function. This network can learn non-linear interactions between task"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2007.02693","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/2007.02693/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":"2007.02693","created_at":"2026-07-05T02:39:02.752482+00:00"},{"alias_kind":"arxiv_version","alias_value":"2007.02693v3","created_at":"2026-07-05T02:39:02.752482+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2007.02693","created_at":"2026-07-05T02:39:02.752482+00:00"},{"alias_kind":"pith_short_12","alias_value":"JZ567H7WDJKX","created_at":"2026-07-05T02:39:02.752482+00:00"},{"alias_kind":"pith_short_16","alias_value":"JZ567H7WDJKX6HNA","created_at":"2026-07-05T02:39:02.752482+00:00"},{"alias_kind":"pith_short_8","alias_value":"JZ567H7W","created_at":"2026-07-05T02:39:02.752482+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.04391","citing_title":"Towards Fair and Robust Face Parsing for Generative AI: A Multi-Objective Approach","ref_index":24,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JZ567H7WDJKX6HNAMKUNCO67CQ","json":"https://pith.science/pith/JZ567H7WDJKX6HNAMKUNCO67CQ.json","graph_json":"https://pith.science/api/pith-number/JZ567H7WDJKX6HNAMKUNCO67CQ/graph.json","events_json":"https://pith.science/api/pith-number/JZ567H7WDJKX6HNAMKUNCO67CQ/events.json","paper":"https://pith.science/paper/JZ567H7W"},"agent_actions":{"view_html":"https://pith.science/pith/JZ567H7WDJKX6HNAMKUNCO67CQ","download_json":"https://pith.science/pith/JZ567H7WDJKX6HNAMKUNCO67CQ.json","view_paper":"https://pith.science/paper/JZ567H7W","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2007.02693&json=true","fetch_graph":"https://pith.science/api/pith-number/JZ567H7WDJKX6HNAMKUNCO67CQ/graph.json","fetch_events":"https://pith.science/api/pith-number/JZ567H7WDJKX6HNAMKUNCO67CQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JZ567H7WDJKX6HNAMKUNCO67CQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JZ567H7WDJKX6HNAMKUNCO67CQ/action/storage_attestation","attest_author":"https://pith.science/pith/JZ567H7WDJKX6HNAMKUNCO67CQ/action/author_attestation","sign_citation":"https://pith.science/pith/JZ567H7WDJKX6HNAMKUNCO67CQ/action/citation_signature","submit_replication":"https://pith.science/pith/JZ567H7WDJKX6HNAMKUNCO67CQ/action/replication_record"}},"created_at":"2026-07-05T02:39:02.752482+00:00","updated_at":"2026-07-05T02:39:02.752482+00:00"}