{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:Z7WI42A6LGTMCSOAAE63RSV6ZX","short_pith_number":"pith:Z7WI42A6","schema_version":"1.0","canonical_sha256":"cfec8e681e59a6c149c0013db8cabecde6defd52571e859488dceafbe9d47ed4","source":{"kind":"arxiv","id":"2409.19460","version":1},"attestation_state":"computed","paper":{"title":"On the universality of neural encodings in CNNs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Brice M\\'enard, Florentin Guth","submitted_at":"2024-09-28T21:30:25Z","abstract_excerpt":"We explore the universality of neural encodings in convolutional neural networks trained on image classification tasks. We develop a procedure to directly compare the learned weights rather than their representations. It is based on a factorization of spatial and channel dimensions and measures the similarity of aligned weight covariances. We show that, for a range of layers of VGG-type networks, the learned eigenvectors appear to be universal across different natural image datasets. Our results suggest the existence of a universal neural encoding for natural images. They explain, at a more fu"},"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":"2409.19460","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-09-28T21:30:25Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"1e02b22449e2fea185e6c30c8c900c569bdfcfb1ebc68909b4071a8404f0fbec","abstract_canon_sha256":"0ce33c4cd8849b3918030518b3f8e49fd0aedc735407bfb5cd3d98f49ec8dad0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:13:15.745937Z","signature_b64":"+ER2+tJ6aLudvYLOkSoE3edURvdmJGD/BSOlTeMjHh3wlxBLY1dfKY+d29WOcXQdnUt9L1jk0gs+ypPwBHRoBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cfec8e681e59a6c149c0013db8cabecde6defd52571e859488dceafbe9d47ed4","last_reissued_at":"2026-07-05T09:13:15.745519Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:13:15.745519Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On the universality of neural encodings in CNNs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Brice M\\'enard, Florentin Guth","submitted_at":"2024-09-28T21:30:25Z","abstract_excerpt":"We explore the universality of neural encodings in convolutional neural networks trained on image classification tasks. We develop a procedure to directly compare the learned weights rather than their representations. It is based on a factorization of spatial and channel dimensions and measures the similarity of aligned weight covariances. We show that, for a range of layers of VGG-type networks, the learned eigenvectors appear to be universal across different natural image datasets. Our results suggest the existence of a universal neural encoding for natural images. They explain, at a more fu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.19460","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/2409.19460/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":"2409.19460","created_at":"2026-07-05T09:13:15.745578+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.19460v1","created_at":"2026-07-05T09:13:15.745578+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.19460","created_at":"2026-07-05T09:13:15.745578+00:00"},{"alias_kind":"pith_short_12","alias_value":"Z7WI42A6LGTM","created_at":"2026-07-05T09:13:15.745578+00:00"},{"alias_kind":"pith_short_16","alias_value":"Z7WI42A6LGTMCSOA","created_at":"2026-07-05T09:13:15.745578+00:00"},{"alias_kind":"pith_short_8","alias_value":"Z7WI42A6","created_at":"2026-07-05T09:13:15.745578+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.05866","citing_title":"Differentially Private Natural Gradient Descent","ref_index":22,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Z7WI42A6LGTMCSOAAE63RSV6ZX","json":"https://pith.science/pith/Z7WI42A6LGTMCSOAAE63RSV6ZX.json","graph_json":"https://pith.science/api/pith-number/Z7WI42A6LGTMCSOAAE63RSV6ZX/graph.json","events_json":"https://pith.science/api/pith-number/Z7WI42A6LGTMCSOAAE63RSV6ZX/events.json","paper":"https://pith.science/paper/Z7WI42A6"},"agent_actions":{"view_html":"https://pith.science/pith/Z7WI42A6LGTMCSOAAE63RSV6ZX","download_json":"https://pith.science/pith/Z7WI42A6LGTMCSOAAE63RSV6ZX.json","view_paper":"https://pith.science/paper/Z7WI42A6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.19460&json=true","fetch_graph":"https://pith.science/api/pith-number/Z7WI42A6LGTMCSOAAE63RSV6ZX/graph.json","fetch_events":"https://pith.science/api/pith-number/Z7WI42A6LGTMCSOAAE63RSV6ZX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Z7WI42A6LGTMCSOAAE63RSV6ZX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Z7WI42A6LGTMCSOAAE63RSV6ZX/action/storage_attestation","attest_author":"https://pith.science/pith/Z7WI42A6LGTMCSOAAE63RSV6ZX/action/author_attestation","sign_citation":"https://pith.science/pith/Z7WI42A6LGTMCSOAAE63RSV6ZX/action/citation_signature","submit_replication":"https://pith.science/pith/Z7WI42A6LGTMCSOAAE63RSV6ZX/action/replication_record"}},"created_at":"2026-07-05T09:13:15.745578+00:00","updated_at":"2026-07-05T09:13:15.745578+00:00"}