{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:G4QD6GS2SHNSSQ2E3LH44AHRYS","short_pith_number":"pith:G4QD6GS2","schema_version":"1.0","canonical_sha256":"37203f1a5a91db294344dacfce00f1c4861f099f684138e34f6978d020c303a2","source":{"kind":"arxiv","id":"1905.09677","version":2},"attestation_state":"computed","paper":{"title":"The role of invariance in spectral complexity-based generalization bounds","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Andreas Loukas, Konstantinos Pitas, Mike Davies, Pierre Vandergheynst","submitted_at":"2019-05-23T14:23:50Z","abstract_excerpt":"Deep convolutional neural networks (CNNs) have been shown to be able to fit a random labeling over data while still being able to generalize well for normal labels. Describing CNN capacity through a posteriori measures of complexity has been recently proposed to tackle this apparent paradox. These complexity measures are usually validated by showing that they correlate empirically with GE; being empirically larger for networks trained on random vs normal labels. Focusing on the case of spectral complexity we investigate theoretically and empirically the insensitivity of the complexity 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":"1905.09677","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-05-23T14:23:50Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"042381f2521cc0752b323ba581be30c4e6f957fa4c6efd8cf4b6b0df67d6eea8","abstract_canon_sha256":"e5aed77762120f4d238cc82bf69aa06a9ee2cff9e614d5b2042d77509132c84d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:46:31.112486Z","signature_b64":"kqtvfeRN4/HlBlqTkpXgi5svZalshAs8Cuq0yPn3yksIHFv1XDFGKTyVc0FBcNS7K7MQze1XJpt0odU6V6wJBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"37203f1a5a91db294344dacfce00f1c4861f099f684138e34f6978d020c303a2","last_reissued_at":"2026-07-05T00:46:31.112077Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:46:31.112077Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The role of invariance in spectral complexity-based generalization bounds","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Andreas Loukas, Konstantinos Pitas, Mike Davies, Pierre Vandergheynst","submitted_at":"2019-05-23T14:23:50Z","abstract_excerpt":"Deep convolutional neural networks (CNNs) have been shown to be able to fit a random labeling over data while still being able to generalize well for normal labels. Describing CNN capacity through a posteriori measures of complexity has been recently proposed to tackle this apparent paradox. These complexity measures are usually validated by showing that they correlate empirically with GE; being empirically larger for networks trained on random vs normal labels. Focusing on the case of spectral complexity we investigate theoretically and empirically the insensitivity of the complexity measure "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1905.09677","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/1905.09677/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":"1905.09677","created_at":"2026-07-05T00:46:31.112143+00:00"},{"alias_kind":"arxiv_version","alias_value":"1905.09677v2","created_at":"2026-07-05T00:46:31.112143+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1905.09677","created_at":"2026-07-05T00:46:31.112143+00:00"},{"alias_kind":"pith_short_12","alias_value":"G4QD6GS2SHNS","created_at":"2026-07-05T00:46:31.112143+00:00"},{"alias_kind":"pith_short_16","alias_value":"G4QD6GS2SHNSSQ2E","created_at":"2026-07-05T00:46:31.112143+00:00"},{"alias_kind":"pith_short_8","alias_value":"G4QD6GS2","created_at":"2026-07-05T00:46:31.112143+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2411.14288","citing_title":"On the Sample Complexity of One Hidden Layer Networks with Equivariance, Locality and Weight Sharing","ref_index":1989,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/G4QD6GS2SHNSSQ2E3LH44AHRYS","json":"https://pith.science/pith/G4QD6GS2SHNSSQ2E3LH44AHRYS.json","graph_json":"https://pith.science/api/pith-number/G4QD6GS2SHNSSQ2E3LH44AHRYS/graph.json","events_json":"https://pith.science/api/pith-number/G4QD6GS2SHNSSQ2E3LH44AHRYS/events.json","paper":"https://pith.science/paper/G4QD6GS2"},"agent_actions":{"view_html":"https://pith.science/pith/G4QD6GS2SHNSSQ2E3LH44AHRYS","download_json":"https://pith.science/pith/G4QD6GS2SHNSSQ2E3LH44AHRYS.json","view_paper":"https://pith.science/paper/G4QD6GS2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1905.09677&json=true","fetch_graph":"https://pith.science/api/pith-number/G4QD6GS2SHNSSQ2E3LH44AHRYS/graph.json","fetch_events":"https://pith.science/api/pith-number/G4QD6GS2SHNSSQ2E3LH44AHRYS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/G4QD6GS2SHNSSQ2E3LH44AHRYS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/G4QD6GS2SHNSSQ2E3LH44AHRYS/action/storage_attestation","attest_author":"https://pith.science/pith/G4QD6GS2SHNSSQ2E3LH44AHRYS/action/author_attestation","sign_citation":"https://pith.science/pith/G4QD6GS2SHNSSQ2E3LH44AHRYS/action/citation_signature","submit_replication":"https://pith.science/pith/G4QD6GS2SHNSSQ2E3LH44AHRYS/action/replication_record"}},"created_at":"2026-07-05T00:46:31.112143+00:00","updated_at":"2026-07-05T00:46:31.112143+00:00"}