{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:SF2BRLDBKKMBRJWE6LL6Y6FKX6","short_pith_number":"pith:SF2BRLDB","schema_version":"1.0","canonical_sha256":"917418ac61529818a6c4f2d7ec78aabf87a0f26831b13c01273f7d2cda9ec5a8","source":{"kind":"arxiv","id":"2106.05121","version":2},"attestation_state":"computed","paper":{"title":"Grounding inductive biases in natural images:invariance stems from variations in data","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Ari S. Morcos, Diane Bouchacourt, Mark Ibrahim","submitted_at":"2021-06-09T14:58:57Z","abstract_excerpt":"To perform well on unseen and potentially out-of-distribution samples, it is desirable for machine learning models to have a predictable response with respect to transformations affecting the factors of variation of the input. Here, we study the relative importance of several types of inductive biases towards such predictable behavior: the choice of data, their augmentations, and model architectures. Invariance is commonly achieved through hand-engineered data augmentation, but do standard data augmentations address transformations that explain variations in real data? While prior work has foc"},"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":"2106.05121","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-06-09T14:58:57Z","cross_cats_sorted":[],"title_canon_sha256":"105427e3205afd4f676e6d1737a692449c0d153142fad780aaa9ccdd219659b8","abstract_canon_sha256":"6933599d44c51b017bdae3e9bb6cf23fb284c95589de0660e4558f0b979da597"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:32:05.149779Z","signature_b64":"tvv40Io487EzlWQtSvcpX3Zd94UuMKJWLwEKSEE44DzqBCrrGsDVUDYnWnAramLx1rPmSwH5/JRqAmwjZK11Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"917418ac61529818a6c4f2d7ec78aabf87a0f26831b13c01273f7d2cda9ec5a8","last_reissued_at":"2026-07-05T03:32:05.149288Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:32:05.149288Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Grounding inductive biases in natural images:invariance stems from variations in data","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Ari S. Morcos, Diane Bouchacourt, Mark Ibrahim","submitted_at":"2021-06-09T14:58:57Z","abstract_excerpt":"To perform well on unseen and potentially out-of-distribution samples, it is desirable for machine learning models to have a predictable response with respect to transformations affecting the factors of variation of the input. Here, we study the relative importance of several types of inductive biases towards such predictable behavior: the choice of data, their augmentations, and model architectures. Invariance is commonly achieved through hand-engineered data augmentation, but do standard data augmentations address transformations that explain variations in real data? While prior work has foc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.05121","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/2106.05121/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":"2106.05121","created_at":"2026-07-05T03:32:05.149356+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.05121v2","created_at":"2026-07-05T03:32:05.149356+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.05121","created_at":"2026-07-05T03:32:05.149356+00:00"},{"alias_kind":"pith_short_12","alias_value":"SF2BRLDBKKMB","created_at":"2026-07-05T03:32:05.149356+00:00"},{"alias_kind":"pith_short_16","alias_value":"SF2BRLDBKKMBRJWE","created_at":"2026-07-05T03:32:05.149356+00:00"},{"alias_kind":"pith_short_8","alias_value":"SF2BRLDB","created_at":"2026-07-05T03:32:05.149356+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SF2BRLDBKKMBRJWE6LL6Y6FKX6","json":"https://pith.science/pith/SF2BRLDBKKMBRJWE6LL6Y6FKX6.json","graph_json":"https://pith.science/api/pith-number/SF2BRLDBKKMBRJWE6LL6Y6FKX6/graph.json","events_json":"https://pith.science/api/pith-number/SF2BRLDBKKMBRJWE6LL6Y6FKX6/events.json","paper":"https://pith.science/paper/SF2BRLDB"},"agent_actions":{"view_html":"https://pith.science/pith/SF2BRLDBKKMBRJWE6LL6Y6FKX6","download_json":"https://pith.science/pith/SF2BRLDBKKMBRJWE6LL6Y6FKX6.json","view_paper":"https://pith.science/paper/SF2BRLDB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.05121&json=true","fetch_graph":"https://pith.science/api/pith-number/SF2BRLDBKKMBRJWE6LL6Y6FKX6/graph.json","fetch_events":"https://pith.science/api/pith-number/SF2BRLDBKKMBRJWE6LL6Y6FKX6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SF2BRLDBKKMBRJWE6LL6Y6FKX6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SF2BRLDBKKMBRJWE6LL6Y6FKX6/action/storage_attestation","attest_author":"https://pith.science/pith/SF2BRLDBKKMBRJWE6LL6Y6FKX6/action/author_attestation","sign_citation":"https://pith.science/pith/SF2BRLDBKKMBRJWE6LL6Y6FKX6/action/citation_signature","submit_replication":"https://pith.science/pith/SF2BRLDBKKMBRJWE6LL6Y6FKX6/action/replication_record"}},"created_at":"2026-07-05T03:32:05.149356+00:00","updated_at":"2026-07-05T03:32:05.149356+00:00"}