{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:TR643GLWHZTKXCBM4ZKOFXXVV6","short_pith_number":"pith:TR643GLW","schema_version":"1.0","canonical_sha256":"9c7dcd99763e66ab882ce654e2def5afbfde5b32194e66e9929c9601143c0cfc","source":{"kind":"arxiv","id":"2010.05761","version":2},"attestation_state":"computed","paper":{"title":"The Risks of Invariant Risk Minimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Andrej Risteski, Elan Rosenfeld, Pradeep Ravikumar","submitted_at":"2020-10-12T14:54:32Z","abstract_excerpt":"Invariant Causal Prediction (Peters et al., 2016) is a technique for out-of-distribution generalization which assumes that some aspects of the data distribution vary across the training set but that the underlying causal mechanisms remain constant. Recently, Arjovsky et al. (2019) proposed Invariant Risk Minimization (IRM), an objective based on this idea for learning deep, invariant features of data which are a complex function of latent variables; many alternatives have subsequently been suggested. However, formal guarantees for all of these works are severely lacking. In this paper, we pres"},"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":"2010.05761","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-10-12T14:54:32Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"f290bfe660db798304a534c4ffad0df14b4ba8d9f9f8031e00790b2181616aa7","abstract_canon_sha256":"f961e9c83c5e966fcb25cf09db90975dafdc8c3c7467d273a9b7c6a6c854fa99"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:26:50.838983Z","signature_b64":"UwwhvKuK05Cz5WbZkEtXbWlQl9LeVwIPsMBZdi64JbBw36PxSYS6kAtft8ipnGDvsA7D+z3GfD3mE4OxuMBrBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9c7dcd99763e66ab882ce654e2def5afbfde5b32194e66e9929c9601143c0cfc","last_reissued_at":"2026-07-05T02:26:50.838470Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:26:50.838470Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Risks of Invariant Risk Minimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Andrej Risteski, Elan Rosenfeld, Pradeep Ravikumar","submitted_at":"2020-10-12T14:54:32Z","abstract_excerpt":"Invariant Causal Prediction (Peters et al., 2016) is a technique for out-of-distribution generalization which assumes that some aspects of the data distribution vary across the training set but that the underlying causal mechanisms remain constant. Recently, Arjovsky et al. (2019) proposed Invariant Risk Minimization (IRM), an objective based on this idea for learning deep, invariant features of data which are a complex function of latent variables; many alternatives have subsequently been suggested. However, formal guarantees for all of these works are severely lacking. In this paper, we pres"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.05761","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/2010.05761/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":"2010.05761","created_at":"2026-07-05T02:26:50.838534+00:00"},{"alias_kind":"arxiv_version","alias_value":"2010.05761v2","created_at":"2026-07-05T02:26:50.838534+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.05761","created_at":"2026-07-05T02:26:50.838534+00:00"},{"alias_kind":"pith_short_12","alias_value":"TR643GLWHZTK","created_at":"2026-07-05T02:26:50.838534+00:00"},{"alias_kind":"pith_short_16","alias_value":"TR643GLWHZTKXCBM","created_at":"2026-07-05T02:26:50.838534+00:00"},{"alias_kind":"pith_short_8","alias_value":"TR643GLW","created_at":"2026-07-05T02:26:50.838534+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.12930","citing_title":"Is Spurious Correlation Removal Always Learnable?","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2606.05025","citing_title":"Invariant Gradient Alignment for Robust Reasoning Distillation","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12945","citing_title":"Separating Shortcut Transition from Cross-Family OOD Failure in a Minimal Model","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12945","citing_title":"Separating Shortcut Transition from Cross-Family OOD Failure in a Minimal Model","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2604.02564","citing_title":"Why Invariance is Not Enough for Biomedical Domain Generalization and How to Fix It","ref_index":50,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06522","citing_title":"Agentic AIs Are the Missing Paradigm for Out-of-Distribution Generalization in Foundation Models","ref_index":36,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TR643GLWHZTKXCBM4ZKOFXXVV6","json":"https://pith.science/pith/TR643GLWHZTKXCBM4ZKOFXXVV6.json","graph_json":"https://pith.science/api/pith-number/TR643GLWHZTKXCBM4ZKOFXXVV6/graph.json","events_json":"https://pith.science/api/pith-number/TR643GLWHZTKXCBM4ZKOFXXVV6/events.json","paper":"https://pith.science/paper/TR643GLW"},"agent_actions":{"view_html":"https://pith.science/pith/TR643GLWHZTKXCBM4ZKOFXXVV6","download_json":"https://pith.science/pith/TR643GLWHZTKXCBM4ZKOFXXVV6.json","view_paper":"https://pith.science/paper/TR643GLW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2010.05761&json=true","fetch_graph":"https://pith.science/api/pith-number/TR643GLWHZTKXCBM4ZKOFXXVV6/graph.json","fetch_events":"https://pith.science/api/pith-number/TR643GLWHZTKXCBM4ZKOFXXVV6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TR643GLWHZTKXCBM4ZKOFXXVV6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TR643GLWHZTKXCBM4ZKOFXXVV6/action/storage_attestation","attest_author":"https://pith.science/pith/TR643GLWHZTKXCBM4ZKOFXXVV6/action/author_attestation","sign_citation":"https://pith.science/pith/TR643GLWHZTKXCBM4ZKOFXXVV6/action/citation_signature","submit_replication":"https://pith.science/pith/TR643GLWHZTKXCBM4ZKOFXXVV6/action/replication_record"}},"created_at":"2026-07-05T02:26:50.838534+00:00","updated_at":"2026-07-05T02:26:50.838534+00:00"}