{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:2DJ7EDCSFS5DGD4TSK6RRXCOIM","short_pith_number":"pith:2DJ7EDCS","schema_version":"1.0","canonical_sha256":"d0d3f20c522cba330f9392bd18dc4e4320c503693cb5a0c706e4552aa5e62860","source":{"kind":"arxiv","id":"2103.12947","version":1},"attestation_state":"computed","paper":{"title":"Meta-Learned Invariant Risk Minimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Inchul Choi, Jun-Hyun Bae, Minho Lee","submitted_at":"2021-03-24T02:52:48Z","abstract_excerpt":"Empirical Risk Minimization (ERM) based machine learning algorithms have suffered from weak generalization performance on data obtained from out-of-distribution (OOD). To address this problem, Invariant Risk Minimization (IRM) objective was suggested to find invariant optimal predictor which is less affected by the changes in data distribution. However, even with such progress, IRMv1, the practical formulation of IRM, still shows performance degradation when there are not enough training data, and even fails to generalize to OOD, if the number of spurious correlations is larger than the number"},"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":"2103.12947","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-03-24T02:52:48Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c2366cbdd17f578e5f4cf4dc7733dedb55e0ae3a1d1c975d130a23735e282354","abstract_canon_sha256":"b417c8ae06358126249b5ea4ae687e0541c3e67dcdd7542f80cf52cdd4ddb84a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:26:04.027719Z","signature_b64":"363o5hjBtufIqQekHliEVdKCWYJlGHlo995599EXqSJbr0SifxFqAzL3SO4ahefZJ4vFMKmDfdVuAEBmsevcAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d0d3f20c522cba330f9392bd18dc4e4320c503693cb5a0c706e4552aa5e62860","last_reissued_at":"2026-07-05T02:26:04.027275Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:26:04.027275Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Meta-Learned Invariant Risk Minimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Inchul Choi, Jun-Hyun Bae, Minho Lee","submitted_at":"2021-03-24T02:52:48Z","abstract_excerpt":"Empirical Risk Minimization (ERM) based machine learning algorithms have suffered from weak generalization performance on data obtained from out-of-distribution (OOD). To address this problem, Invariant Risk Minimization (IRM) objective was suggested to find invariant optimal predictor which is less affected by the changes in data distribution. However, even with such progress, IRMv1, the practical formulation of IRM, still shows performance degradation when there are not enough training data, and even fails to generalize to OOD, if the number of spurious correlations is larger than the number"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2103.12947","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/2103.12947/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":"2103.12947","created_at":"2026-07-05T02:26:04.027371+00:00"},{"alias_kind":"arxiv_version","alias_value":"2103.12947v1","created_at":"2026-07-05T02:26:04.027371+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2103.12947","created_at":"2026-07-05T02:26:04.027371+00:00"},{"alias_kind":"pith_short_12","alias_value":"2DJ7EDCSFS5D","created_at":"2026-07-05T02:26:04.027371+00:00"},{"alias_kind":"pith_short_16","alias_value":"2DJ7EDCSFS5DGD4T","created_at":"2026-07-05T02:26:04.027371+00:00"},{"alias_kind":"pith_short_8","alias_value":"2DJ7EDCS","created_at":"2026-07-05T02:26:04.027371+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.11756","citing_title":"Causal Effect Identification in Heterogeneous Environments from Higher-Order Moments","ref_index":3,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2DJ7EDCSFS5DGD4TSK6RRXCOIM","json":"https://pith.science/pith/2DJ7EDCSFS5DGD4TSK6RRXCOIM.json","graph_json":"https://pith.science/api/pith-number/2DJ7EDCSFS5DGD4TSK6RRXCOIM/graph.json","events_json":"https://pith.science/api/pith-number/2DJ7EDCSFS5DGD4TSK6RRXCOIM/events.json","paper":"https://pith.science/paper/2DJ7EDCS"},"agent_actions":{"view_html":"https://pith.science/pith/2DJ7EDCSFS5DGD4TSK6RRXCOIM","download_json":"https://pith.science/pith/2DJ7EDCSFS5DGD4TSK6RRXCOIM.json","view_paper":"https://pith.science/paper/2DJ7EDCS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2103.12947&json=true","fetch_graph":"https://pith.science/api/pith-number/2DJ7EDCSFS5DGD4TSK6RRXCOIM/graph.json","fetch_events":"https://pith.science/api/pith-number/2DJ7EDCSFS5DGD4TSK6RRXCOIM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2DJ7EDCSFS5DGD4TSK6RRXCOIM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2DJ7EDCSFS5DGD4TSK6RRXCOIM/action/storage_attestation","attest_author":"https://pith.science/pith/2DJ7EDCSFS5DGD4TSK6RRXCOIM/action/author_attestation","sign_citation":"https://pith.science/pith/2DJ7EDCSFS5DGD4TSK6RRXCOIM/action/citation_signature","submit_replication":"https://pith.science/pith/2DJ7EDCSFS5DGD4TSK6RRXCOIM/action/replication_record"}},"created_at":"2026-07-05T02:26:04.027371+00:00","updated_at":"2026-07-05T02:26:04.027371+00:00"}