{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:KPCFQET4SLETJCN4JSMYDZYTC2","short_pith_number":"pith:KPCFQET4","schema_version":"1.0","canonical_sha256":"53c458127c92c93489bc4c9981e71316a62e226242ca20da2cadd3e900f1260b","source":{"kind":"arxiv","id":"2303.17765","version":5},"attestation_state":"computed","paper":{"title":"Learning from Similar Linear Representations: Adaptivity, Minimaxity, and Robustness","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Yang Feng, Ye Tian, Yuqi Gu","submitted_at":"2023-03-31T01:56:13Z","abstract_excerpt":"Representation multi-task learning (MTL) has achieved tremendous success in practice. However, the theoretical understanding of these methods is still lacking. Most existing theoretical works focus on cases where all tasks share the same representation, and claim that MTL almost always improves performance. Nevertheless, as the number of tasks grows, assuming all tasks share the same representation is unrealistic. Furthermore, empirical findings often indicate that a shared representation does not necessarily improve single-task learning performance. In this paper, we aim to understand how to "},"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":"2303.17765","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2023-03-31T01:56:13Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"0aeb7fa4de975055fae4f8ba970ae14888e3a229d41c3d9f604e15a2adb978e8","abstract_canon_sha256":"07a78eeb108f2d5e1051955ae975036f8556a48f7f70ce21bd2d9506e8876c18"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:32:12.626353Z","signature_b64":"TeEjR1Ah4mZVMWJ8VzfsURiQvIi90UTsCPtDaZCgh+8A02R+DV0vUqS/A3z19UvQiy8yiujCpt+cE9xpAquHBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"53c458127c92c93489bc4c9981e71316a62e226242ca20da2cadd3e900f1260b","last_reissued_at":"2026-07-05T11:32:12.625743Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:32:12.625743Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning from Similar Linear Representations: Adaptivity, Minimaxity, and Robustness","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Yang Feng, Ye Tian, Yuqi Gu","submitted_at":"2023-03-31T01:56:13Z","abstract_excerpt":"Representation multi-task learning (MTL) has achieved tremendous success in practice. However, the theoretical understanding of these methods is still lacking. Most existing theoretical works focus on cases where all tasks share the same representation, and claim that MTL almost always improves performance. Nevertheless, as the number of tasks grows, assuming all tasks share the same representation is unrealistic. Furthermore, empirical findings often indicate that a shared representation does not necessarily improve single-task learning performance. In this paper, we aim to understand how to "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.17765","kind":"arxiv","version":5},"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/2303.17765/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":"2303.17765","created_at":"2026-07-05T11:32:12.625798+00:00"},{"alias_kind":"arxiv_version","alias_value":"2303.17765v5","created_at":"2026-07-05T11:32:12.625798+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.17765","created_at":"2026-07-05T11:32:12.625798+00:00"},{"alias_kind":"pith_short_12","alias_value":"KPCFQET4SLET","created_at":"2026-07-05T11:32:12.625798+00:00"},{"alias_kind":"pith_short_16","alias_value":"KPCFQET4SLETJCN4","created_at":"2026-07-05T11:32:12.625798+00:00"},{"alias_kind":"pith_short_8","alias_value":"KPCFQET4","created_at":"2026-07-05T11:32:12.625798+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.08981","citing_title":"Divide-and-shrink: An efficient and heterogeneity-agnostic approach for transfer estimation using summary statistics","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2409.02708","citing_title":"Few-shot Multi-Task Learning of Linear Invariant Features with Meta Subspace Pursuit","ref_index":51,"is_internal_anchor":false},{"citing_arxiv_id":"2604.25845","citing_title":"Model-agnostic information transfer and fusion for classification with label noise","ref_index":4,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KPCFQET4SLETJCN4JSMYDZYTC2","json":"https://pith.science/pith/KPCFQET4SLETJCN4JSMYDZYTC2.json","graph_json":"https://pith.science/api/pith-number/KPCFQET4SLETJCN4JSMYDZYTC2/graph.json","events_json":"https://pith.science/api/pith-number/KPCFQET4SLETJCN4JSMYDZYTC2/events.json","paper":"https://pith.science/paper/KPCFQET4"},"agent_actions":{"view_html":"https://pith.science/pith/KPCFQET4SLETJCN4JSMYDZYTC2","download_json":"https://pith.science/pith/KPCFQET4SLETJCN4JSMYDZYTC2.json","view_paper":"https://pith.science/paper/KPCFQET4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2303.17765&json=true","fetch_graph":"https://pith.science/api/pith-number/KPCFQET4SLETJCN4JSMYDZYTC2/graph.json","fetch_events":"https://pith.science/api/pith-number/KPCFQET4SLETJCN4JSMYDZYTC2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KPCFQET4SLETJCN4JSMYDZYTC2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KPCFQET4SLETJCN4JSMYDZYTC2/action/storage_attestation","attest_author":"https://pith.science/pith/KPCFQET4SLETJCN4JSMYDZYTC2/action/author_attestation","sign_citation":"https://pith.science/pith/KPCFQET4SLETJCN4JSMYDZYTC2/action/citation_signature","submit_replication":"https://pith.science/pith/KPCFQET4SLETJCN4JSMYDZYTC2/action/replication_record"}},"created_at":"2026-07-05T11:32:12.625798+00:00","updated_at":"2026-07-05T11:32:12.625798+00:00"}