{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:FLEBEIX4IRGNYYOETNUILH3H32","short_pith_number":"pith:FLEBEIX4","schema_version":"1.0","canonical_sha256":"2ac81222fc444cdc61c49b68859f67de8651bafb812146094cbe611b316699ab","source":{"kind":"arxiv","id":"1711.01530","version":2},"attestation_state":"computed","paper":{"title":"Fisher-Rao Metric, Geometry, and Complexity of Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Alexander Rakhlin, James Stokes, Tengyuan Liang, Tomaso Poggio","submitted_at":"2017-11-05T04:32:59Z","abstract_excerpt":"We study the relationship between geometry and capacity measures for deep neural networks from an invariance viewpoint. We introduce a new notion of capacity --- the Fisher-Rao norm --- that possesses desirable invariance properties and is motivated by Information Geometry. We discover an analytical characterization of the new capacity measure, through which we establish norm-comparison inequalities and further show that the new measure serves as an umbrella for several existing norm-based complexity measures. We discuss upper bounds on the generalization error induced by the proposed measure."},"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":"1711.01530","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-11-05T04:32:59Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"ed8e31f17b95ac740916dc8a69ec0a756fe47b38ba936522b001ef7ceee13429","abstract_canon_sha256":"114f614c5ddd8f1beba8e8541d739f47a0f984fc31dda3375447ea0163ca9e0c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:21:44.534276Z","signature_b64":"sXtnbJ4cyFSkVVazIfB27QB3VyqwRsg7N6vK5i9q1rIqznKHfo77eccDVNh6Jfa0h4fI6/SYv+HlkzjUWFpBAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2ac81222fc444cdc61c49b68859f67de8651bafb812146094cbe611b316699ab","last_reissued_at":"2026-07-05T01:21:44.533772Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:21:44.533772Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Fisher-Rao Metric, Geometry, and Complexity of Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Alexander Rakhlin, James Stokes, Tengyuan Liang, Tomaso Poggio","submitted_at":"2017-11-05T04:32:59Z","abstract_excerpt":"We study the relationship between geometry and capacity measures for deep neural networks from an invariance viewpoint. We introduce a new notion of capacity --- the Fisher-Rao norm --- that possesses desirable invariance properties and is motivated by Information Geometry. We discover an analytical characterization of the new capacity measure, through which we establish norm-comparison inequalities and further show that the new measure serves as an umbrella for several existing norm-based complexity measures. We discuss upper bounds on the generalization error induced by the proposed measure."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1711.01530","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/1711.01530/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":"1711.01530","created_at":"2026-07-05T01:21:44.533836+00:00"},{"alias_kind":"arxiv_version","alias_value":"1711.01530v2","created_at":"2026-07-05T01:21:44.533836+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1711.01530","created_at":"2026-07-05T01:21:44.533836+00:00"},{"alias_kind":"pith_short_12","alias_value":"FLEBEIX4IRGN","created_at":"2026-07-05T01:21:44.533836+00:00"},{"alias_kind":"pith_short_16","alias_value":"FLEBEIX4IRGNYYOE","created_at":"2026-07-05T01:21:44.533836+00:00"},{"alias_kind":"pith_short_8","alias_value":"FLEBEIX4","created_at":"2026-07-05T01:21:44.533836+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"1907.06572","citing_title":"Deep network as memory space: complexity, generalization, disentangled representation and interpretability","ref_index":22,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FLEBEIX4IRGNYYOETNUILH3H32","json":"https://pith.science/pith/FLEBEIX4IRGNYYOETNUILH3H32.json","graph_json":"https://pith.science/api/pith-number/FLEBEIX4IRGNYYOETNUILH3H32/graph.json","events_json":"https://pith.science/api/pith-number/FLEBEIX4IRGNYYOETNUILH3H32/events.json","paper":"https://pith.science/paper/FLEBEIX4"},"agent_actions":{"view_html":"https://pith.science/pith/FLEBEIX4IRGNYYOETNUILH3H32","download_json":"https://pith.science/pith/FLEBEIX4IRGNYYOETNUILH3H32.json","view_paper":"https://pith.science/paper/FLEBEIX4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1711.01530&json=true","fetch_graph":"https://pith.science/api/pith-number/FLEBEIX4IRGNYYOETNUILH3H32/graph.json","fetch_events":"https://pith.science/api/pith-number/FLEBEIX4IRGNYYOETNUILH3H32/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FLEBEIX4IRGNYYOETNUILH3H32/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FLEBEIX4IRGNYYOETNUILH3H32/action/storage_attestation","attest_author":"https://pith.science/pith/FLEBEIX4IRGNYYOETNUILH3H32/action/author_attestation","sign_citation":"https://pith.science/pith/FLEBEIX4IRGNYYOETNUILH3H32/action/citation_signature","submit_replication":"https://pith.science/pith/FLEBEIX4IRGNYYOETNUILH3H32/action/replication_record"}},"created_at":"2026-07-05T01:21:44.533836+00:00","updated_at":"2026-07-05T01:21:44.533836+00:00"}