{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:VT47ORQQZ5XDGDH7W77RPP4BDB","short_pith_number":"pith:VT47ORQQ","canonical_record":{"source":{"id":"1808.04295","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-08-13T15:40:41Z","cross_cats_sorted":["cs.AI","math.OC","math.ST","stat.ML","stat.TH"],"title_canon_sha256":"eb0a34147a46c5e44668ad3de40bfaac0a0673d4025d1bf03c5ea005058e8c64","abstract_canon_sha256":"ade06a957ac648d4c55870a59bbeb8b157c4f189806b0990cb639ff2b13ad872"},"schema_version":"1.0"},"canonical_sha256":"acf9f74610cf6e330cffb7ff17bf81185dbb9cf98f4261a7997d61684d64a411","source":{"kind":"arxiv","id":"1808.04295","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1808.04295","created_at":"2026-05-17T23:59:38Z"},{"alias_kind":"arxiv_version","alias_value":"1808.04295v4","created_at":"2026-05-17T23:59:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1808.04295","created_at":"2026-05-17T23:59:38Z"},{"alias_kind":"pith_short_12","alias_value":"VT47ORQQZ5XD","created_at":"2026-05-18T12:32:59Z"},{"alias_kind":"pith_short_16","alias_value":"VT47ORQQZ5XDGDH7","created_at":"2026-05-18T12:32:59Z"},{"alias_kind":"pith_short_8","alias_value":"VT47ORQQ","created_at":"2026-05-18T12:32:59Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:VT47ORQQZ5XDGDH7W77RPP4BDB","target":"record","payload":{"canonical_record":{"source":{"id":"1808.04295","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-08-13T15:40:41Z","cross_cats_sorted":["cs.AI","math.OC","math.ST","stat.ML","stat.TH"],"title_canon_sha256":"eb0a34147a46c5e44668ad3de40bfaac0a0673d4025d1bf03c5ea005058e8c64","abstract_canon_sha256":"ade06a957ac648d4c55870a59bbeb8b157c4f189806b0990cb639ff2b13ad872"},"schema_version":"1.0"},"canonical_sha256":"acf9f74610cf6e330cffb7ff17bf81185dbb9cf98f4261a7997d61684d64a411","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:59:38.457248Z","signature_b64":"hPAnvtiY0EwiRwmRwz1gEwB0gG96XVde0oiXpRvTnJlz91USGOFFi/ZcJAm0psNlc0dNfwj7v2QaIlVNqCXRAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"acf9f74610cf6e330cffb7ff17bf81185dbb9cf98f4261a7997d61684d64a411","last_reissued_at":"2026-05-17T23:59:38.456614Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:59:38.456614Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1808.04295","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:59:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ucsn36mffmo1M2ehlgVa0asJL6ADLfdRVWlakT3O1y3m+dtbBx9ZXTsjX9IqRfjWXOKEySKVRc/YM0yYbQrRAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-10T00:59:29.450509Z"},"content_sha256":"4c7b2025fde8cb7343ab531762b4b56d9cbf9ae9d7bad429f8b74fcdc75c20d7","schema_version":"1.0","event_id":"sha256:4c7b2025fde8cb7343ab531762b4b56d9cbf9ae9d7bad429f8b74fcdc75c20d7"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:VT47ORQQZ5XDGDH7W77RPP4BDB","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Understanding training and generalization in deep learning by Fourier analysis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","math.OC","math.ST","stat.ML","stat.TH"],"primary_cat":"cs.LG","authors_text":"Zhiqin John Xu","submitted_at":"2018-08-13T15:40:41Z","abstract_excerpt":"Background: It is still an open research area to theoretically understand why Deep Neural Networks (DNNs)---equipped with many more parameters than training data and trained by (stochastic) gradient-based methods---often achieve remarkably low generalization error. Contribution: We study DNN training by Fourier analysis. Our theoretical framework explains: i) DNN with (stochastic) gradient-based methods often endows low-frequency components of the target function with a higher priority during the training; ii) Small initialization leads to good generalization ability of DNN while preserving th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1808.04295","kind":"arxiv","version":4},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:59:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"A5l+a43k6DFqL8eYW5JKnPR6+wrOKS0L1lxSFtCjcr3Q27O2eM7fSnfprz7Y+8pLRerYGgzRdixztRUkorDOCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-10T00:59:29.451204Z"},"content_sha256":"a8aa4d3d9e2ba145fd9b64327aebdb0fd4bc53115594bbbd6b1a74deb51a835b","schema_version":"1.0","event_id":"sha256:a8aa4d3d9e2ba145fd9b64327aebdb0fd4bc53115594bbbd6b1a74deb51a835b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/VT47ORQQZ5XDGDH7W77RPP4BDB/bundle.json","state_url":"https://pith.science/pith/VT47ORQQZ5XDGDH7W77RPP4BDB/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/VT47ORQQZ5XDGDH7W77RPP4BDB/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-06-10T00:59:29Z","links":{"resolver":"https://pith.science/pith/VT47ORQQZ5XDGDH7W77RPP4BDB","bundle":"https://pith.science/pith/VT47ORQQZ5XDGDH7W77RPP4BDB/bundle.json","state":"https://pith.science/pith/VT47ORQQZ5XDGDH7W77RPP4BDB/state.json","well_known_bundle":"https://pith.science/.well-known/pith/VT47ORQQZ5XDGDH7W77RPP4BDB/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:VT47ORQQZ5XDGDH7W77RPP4BDB","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"ade06a957ac648d4c55870a59bbeb8b157c4f189806b0990cb639ff2b13ad872","cross_cats_sorted":["cs.AI","math.OC","math.ST","stat.ML","stat.TH"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-08-13T15:40:41Z","title_canon_sha256":"eb0a34147a46c5e44668ad3de40bfaac0a0673d4025d1bf03c5ea005058e8c64"},"schema_version":"1.0","source":{"id":"1808.04295","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1808.04295","created_at":"2026-05-17T23:59:38Z"},{"alias_kind":"arxiv_version","alias_value":"1808.04295v4","created_at":"2026-05-17T23:59:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1808.04295","created_at":"2026-05-17T23:59:38Z"},{"alias_kind":"pith_short_12","alias_value":"VT47ORQQZ5XD","created_at":"2026-05-18T12:32:59Z"},{"alias_kind":"pith_short_16","alias_value":"VT47ORQQZ5XDGDH7","created_at":"2026-05-18T12:32:59Z"},{"alias_kind":"pith_short_8","alias_value":"VT47ORQQ","created_at":"2026-05-18T12:32:59Z"}],"graph_snapshots":[{"event_id":"sha256:a8aa4d3d9e2ba145fd9b64327aebdb0fd4bc53115594bbbd6b1a74deb51a835b","target":"graph","created_at":"2026-05-17T23:59:38Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"Background: It is still an open research area to theoretically understand why Deep Neural Networks (DNNs)---equipped with many more parameters than training data and trained by (stochastic) gradient-based methods---often achieve remarkably low generalization error. Contribution: We study DNN training by Fourier analysis. Our theoretical framework explains: i) DNN with (stochastic) gradient-based methods often endows low-frequency components of the target function with a higher priority during the training; ii) Small initialization leads to good generalization ability of DNN while preserving th","authors_text":"Zhiqin John Xu","cross_cats":["cs.AI","math.OC","math.ST","stat.ML","stat.TH"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-08-13T15:40:41Z","title":"Understanding training and generalization in deep learning by Fourier analysis"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1808.04295","kind":"arxiv","version":4},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:4c7b2025fde8cb7343ab531762b4b56d9cbf9ae9d7bad429f8b74fcdc75c20d7","target":"record","created_at":"2026-05-17T23:59:38Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"ade06a957ac648d4c55870a59bbeb8b157c4f189806b0990cb639ff2b13ad872","cross_cats_sorted":["cs.AI","math.OC","math.ST","stat.ML","stat.TH"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-08-13T15:40:41Z","title_canon_sha256":"eb0a34147a46c5e44668ad3de40bfaac0a0673d4025d1bf03c5ea005058e8c64"},"schema_version":"1.0","source":{"id":"1808.04295","kind":"arxiv","version":4}},"canonical_sha256":"acf9f74610cf6e330cffb7ff17bf81185dbb9cf98f4261a7997d61684d64a411","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"acf9f74610cf6e330cffb7ff17bf81185dbb9cf98f4261a7997d61684d64a411","first_computed_at":"2026-05-17T23:59:38.456614Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:59:38.456614Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"hPAnvtiY0EwiRwmRwz1gEwB0gG96XVde0oiXpRvTnJlz91USGOFFi/ZcJAm0psNlc0dNfwj7v2QaIlVNqCXRAg==","signature_status":"signed_v1","signed_at":"2026-05-17T23:59:38.457248Z","signed_message":"canonical_sha256_bytes"},"source_id":"1808.04295","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4c7b2025fde8cb7343ab531762b4b56d9cbf9ae9d7bad429f8b74fcdc75c20d7","sha256:a8aa4d3d9e2ba145fd9b64327aebdb0fd4bc53115594bbbd6b1a74deb51a835b"],"state_sha256":"9ab89831957a9773a0bc307cd62a9b438c354d01e512dde852b6c4cb2b37a383"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PsvADziOY6jtqIOnyn2Mmah+qYEfLKF+vgm3O8bXVf9A8aFZmEIr9FEgHYn0xWQHwN2XSJ315stGQIKYXXnaDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-10T00:59:29.455604Z","bundle_sha256":"17bb6bebe7298329fe960575ca465f7402397f529076a6a3cd9974c6a6c6d530"}}