{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:EFRA6FOECG7663B4QSKNZDWAB5","short_pith_number":"pith:EFRA6FOE","schema_version":"1.0","canonical_sha256":"21620f15c411bfef6c3c8494dc8ec00f5f77b01b69cde0ffd45e14479d7f0cf1","source":{"kind":"arxiv","id":"2205.14082","version":2},"attestation_state":"computed","paper":{"title":"AANG: Automating Auxiliary Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Ameet Talwalkar, Graham Neubig, Lucio M. Dery, Mikhail Khodak, Paul Michel","submitted_at":"2022-05-27T16:32:28Z","abstract_excerpt":"Auxiliary objectives, supplementary learning signals that are introduced to help aid learning on data-starved or highly complex end-tasks, are commonplace in machine learning. Whilst much work has been done to formulate useful auxiliary objectives, their construction is still an art which proceeds by slow and tedious hand-design. Intuition for how and when these objectives improve end-task performance has also had limited theoretical backing. In this work, we present an approach for automatically generating a suite of auxiliary objectives. We achieve this by deconstructing existing objectives "},"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":"2205.14082","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-05-27T16:32:28Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"97d2f4832a456ebc60dfd0b283884b4e1bd65411207cdbb1f8c8420d1e184100","abstract_canon_sha256":"9321383fde8db285113efd6c811b11a6832cc6750bfbd37de87350e8bc1e8663"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:46:15.875678Z","signature_b64":"3ZJCGFBGTq83WU7ZHI9IkWqXVL8wr1EHeAt5T50xYocCFg0ohdLk7VcHjVZmVEUjym/3LtmCT1ighUZp+RdnAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"21620f15c411bfef6c3c8494dc8ec00f5f77b01b69cde0ffd45e14479d7f0cf1","last_reissued_at":"2026-07-05T05:46:15.875307Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:46:15.875307Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AANG: Automating Auxiliary Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Ameet Talwalkar, Graham Neubig, Lucio M. Dery, Mikhail Khodak, Paul Michel","submitted_at":"2022-05-27T16:32:28Z","abstract_excerpt":"Auxiliary objectives, supplementary learning signals that are introduced to help aid learning on data-starved or highly complex end-tasks, are commonplace in machine learning. Whilst much work has been done to formulate useful auxiliary objectives, their construction is still an art which proceeds by slow and tedious hand-design. Intuition for how and when these objectives improve end-task performance has also had limited theoretical backing. In this work, we present an approach for automatically generating a suite of auxiliary objectives. We achieve this by deconstructing existing objectives "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.14082","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2205.14082/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":"2205.14082","created_at":"2026-07-05T05:46:15.875375+00:00"},{"alias_kind":"arxiv_version","alias_value":"2205.14082v2","created_at":"2026-07-05T05:46:15.875375+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.14082","created_at":"2026-07-05T05:46:15.875375+00:00"},{"alias_kind":"pith_short_12","alias_value":"EFRA6FOECG76","created_at":"2026-07-05T05:46:15.875375+00:00"},{"alias_kind":"pith_short_16","alias_value":"EFRA6FOECG7663B4","created_at":"2026-07-05T05:46:15.875375+00:00"},{"alias_kind":"pith_short_8","alias_value":"EFRA6FOE","created_at":"2026-07-05T05:46:15.875375+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.12824","citing_title":"Enhancing Monocular Depth Estimation with Multi-Source Auxiliary Tasks","ref_index":17,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EFRA6FOECG7663B4QSKNZDWAB5","json":"https://pith.science/pith/EFRA6FOECG7663B4QSKNZDWAB5.json","graph_json":"https://pith.science/api/pith-number/EFRA6FOECG7663B4QSKNZDWAB5/graph.json","events_json":"https://pith.science/api/pith-number/EFRA6FOECG7663B4QSKNZDWAB5/events.json","paper":"https://pith.science/paper/EFRA6FOE"},"agent_actions":{"view_html":"https://pith.science/pith/EFRA6FOECG7663B4QSKNZDWAB5","download_json":"https://pith.science/pith/EFRA6FOECG7663B4QSKNZDWAB5.json","view_paper":"https://pith.science/paper/EFRA6FOE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2205.14082&json=true","fetch_graph":"https://pith.science/api/pith-number/EFRA6FOECG7663B4QSKNZDWAB5/graph.json","fetch_events":"https://pith.science/api/pith-number/EFRA6FOECG7663B4QSKNZDWAB5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EFRA6FOECG7663B4QSKNZDWAB5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EFRA6FOECG7663B4QSKNZDWAB5/action/storage_attestation","attest_author":"https://pith.science/pith/EFRA6FOECG7663B4QSKNZDWAB5/action/author_attestation","sign_citation":"https://pith.science/pith/EFRA6FOECG7663B4QSKNZDWAB5/action/citation_signature","submit_replication":"https://pith.science/pith/EFRA6FOECG7663B4QSKNZDWAB5/action/replication_record"}},"created_at":"2026-07-05T05:46:15.875375+00:00","updated_at":"2026-07-05T05:46:15.875375+00:00"}