{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:WI2Z3SEHDBROVYA676KWV7UDDM","short_pith_number":"pith:WI2Z3SEH","schema_version":"1.0","canonical_sha256":"b2359dc8871862eae01eff956afe831b3a919411b28aede8886d5889f37a0795","source":{"kind":"arxiv","id":"2212.07346","version":3},"attestation_state":"computed","paper":{"title":"Learning useful representations for shifting tasks and distributions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Jianyu Zhang, L\\'eon Bottou","submitted_at":"2022-12-14T17:17:10Z","abstract_excerpt":"Does the dominant approach to learn representations (as a side effect of optimizing an expected cost for a single training distribution) remain a good approach when we are dealing with multiple distributions? Our thesis is that such scenarios are better served by representations that are richer than those obtained with a single optimization episode. We support this thesis with simple theoretical arguments and with experiments utilizing an apparently na\\\"{\\i}ve ensembling technique: concatenating the representations obtained from multiple training episodes using the same data, model, algorithm,"},"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":"2212.07346","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-12-14T17:17:10Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"8f32f580f23eba915e5f4901f1d0b2dc9c40081d52770306ff2b2718ff1cdb9c","abstract_canon_sha256":"2938a292a4ce6f1eb185f3e0608e389263f37b2f501e1dfe51f435be91a8d2fc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:36:10.808897Z","signature_b64":"dpAgUtWIIxtqgRtyzxP2cI/JK2Ghy9uLv2PuusJOzXPJ/zeCtcnAL5P4KII4aLutcpqs5QwERdMvN+q7XIktAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b2359dc8871862eae01eff956afe831b3a919411b28aede8886d5889f37a0795","last_reissued_at":"2026-07-05T06:36:10.808521Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:36:10.808521Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning useful representations for shifting tasks and distributions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Jianyu Zhang, L\\'eon Bottou","submitted_at":"2022-12-14T17:17:10Z","abstract_excerpt":"Does the dominant approach to learn representations (as a side effect of optimizing an expected cost for a single training distribution) remain a good approach when we are dealing with multiple distributions? Our thesis is that such scenarios are better served by representations that are richer than those obtained with a single optimization episode. We support this thesis with simple theoretical arguments and with experiments utilizing an apparently na\\\"{\\i}ve ensembling technique: concatenating the representations obtained from multiple training episodes using the same data, model, algorithm,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2212.07346","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/2212.07346/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":"2212.07346","created_at":"2026-07-05T06:36:10.808581+00:00"},{"alias_kind":"arxiv_version","alias_value":"2212.07346v3","created_at":"2026-07-05T06:36:10.808581+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2212.07346","created_at":"2026-07-05T06:36:10.808581+00:00"},{"alias_kind":"pith_short_12","alias_value":"WI2Z3SEHDBRO","created_at":"2026-07-05T06:36:10.808581+00:00"},{"alias_kind":"pith_short_16","alias_value":"WI2Z3SEHDBROVYA6","created_at":"2026-07-05T06:36:10.808581+00:00"},{"alias_kind":"pith_short_8","alias_value":"WI2Z3SEH","created_at":"2026-07-05T06:36:10.808581+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.12226","citing_title":"Learning Causality for Modern Machine Learning","ref_index":72,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WI2Z3SEHDBROVYA676KWV7UDDM","json":"https://pith.science/pith/WI2Z3SEHDBROVYA676KWV7UDDM.json","graph_json":"https://pith.science/api/pith-number/WI2Z3SEHDBROVYA676KWV7UDDM/graph.json","events_json":"https://pith.science/api/pith-number/WI2Z3SEHDBROVYA676KWV7UDDM/events.json","paper":"https://pith.science/paper/WI2Z3SEH"},"agent_actions":{"view_html":"https://pith.science/pith/WI2Z3SEHDBROVYA676KWV7UDDM","download_json":"https://pith.science/pith/WI2Z3SEHDBROVYA676KWV7UDDM.json","view_paper":"https://pith.science/paper/WI2Z3SEH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2212.07346&json=true","fetch_graph":"https://pith.science/api/pith-number/WI2Z3SEHDBROVYA676KWV7UDDM/graph.json","fetch_events":"https://pith.science/api/pith-number/WI2Z3SEHDBROVYA676KWV7UDDM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WI2Z3SEHDBROVYA676KWV7UDDM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WI2Z3SEHDBROVYA676KWV7UDDM/action/storage_attestation","attest_author":"https://pith.science/pith/WI2Z3SEHDBROVYA676KWV7UDDM/action/author_attestation","sign_citation":"https://pith.science/pith/WI2Z3SEHDBROVYA676KWV7UDDM/action/citation_signature","submit_replication":"https://pith.science/pith/WI2Z3SEHDBROVYA676KWV7UDDM/action/replication_record"}},"created_at":"2026-07-05T06:36:10.808581+00:00","updated_at":"2026-07-05T06:36:10.808581+00:00"}