{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:HE27HZUSH2OKMTKG5RQ43FF26T","short_pith_number":"pith:HE27HZUS","canonical_record":{"source":{"id":"2208.04924","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-08-09T17:40:57Z","cross_cats_sorted":[],"title_canon_sha256":"97db8ce9aaa3e2cd73797e57d0a0ceaa685eeb4e16870ac46717cbff5ca72de4","abstract_canon_sha256":"cf38ddbf61c9000dd5c866ff06d0e60cc9af214df16c8ff3c6c4b92db79d70a0"},"schema_version":"1.0"},"canonical_sha256":"3935f3e6923e9ca64d46ec61cd94baf4c2c21f546c74618f0f193ac979ea9a49","source":{"kind":"arxiv","id":"2208.04924","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2208.04924","created_at":"2026-07-05T04:55:06Z"},{"alias_kind":"arxiv_version","alias_value":"2208.04924v3","created_at":"2026-07-05T04:55:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2208.04924","created_at":"2026-07-05T04:55:06Z"},{"alias_kind":"pith_short_12","alias_value":"HE27HZUSH2OK","created_at":"2026-07-05T04:55:06Z"},{"alias_kind":"pith_short_16","alias_value":"HE27HZUSH2OKMTKG","created_at":"2026-07-05T04:55:06Z"},{"alias_kind":"pith_short_8","alias_value":"HE27HZUS","created_at":"2026-07-05T04:55:06Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:HE27HZUSH2OKMTKG5RQ43FF26T","target":"record","payload":{"canonical_record":{"source":{"id":"2208.04924","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-08-09T17:40:57Z","cross_cats_sorted":[],"title_canon_sha256":"97db8ce9aaa3e2cd73797e57d0a0ceaa685eeb4e16870ac46717cbff5ca72de4","abstract_canon_sha256":"cf38ddbf61c9000dd5c866ff06d0e60cc9af214df16c8ff3c6c4b92db79d70a0"},"schema_version":"1.0"},"canonical_sha256":"3935f3e6923e9ca64d46ec61cd94baf4c2c21f546c74618f0f193ac979ea9a49","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:55:06.362651Z","signature_b64":"SIaLv4KrutJ64u/lPE2GdN5Xoef871dWDrdbYzXO/j69IqjZZCfdIx1eOqfTYNDFDLpxguNl3wh2Up+P2xG4Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3935f3e6923e9ca64d46ec61cd94baf4c2c21f546c74618f0f193ac979ea9a49","last_reissued_at":"2026-07-05T04:55:06.362107Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:55:06.362107Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2208.04924","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T04:55:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"iZxMx/+F6nqHsXY7C9xcKY3JRh0d2XpsW2H3vmniwCbcGas9E0fH7Tq/vCaedvgDlxczXPc/d9xqxRwW9bCUBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T06:32:20.938110Z"},"content_sha256":"fa281fc338baca358a9e2e4214d80e4da58d26af0faf826cdf4ed3f7a8ed1db2","schema_version":"1.0","event_id":"sha256:fa281fc338baca358a9e2e4214d80e4da58d26af0faf826cdf4ed3f7a8ed1db2"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:HE27HZUSH2OKMTKG5RQ43FF26T","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"On the Activation Function Dependence of the Spectral Bias of Neural Networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Jinchao Xu, Jonathan W. Siegel, Qingguo Hong, Qinyang Tan","submitted_at":"2022-08-09T17:40:57Z","abstract_excerpt":"Neural networks are universal function approximators which are known to generalize well despite being dramatically overparameterized. We study this phenomenon from the point of view of the spectral bias of neural networks. Our contributions are two-fold. First, we provide a theoretical explanation for the spectral bias of ReLU neural networks by leveraging connections with the theory of finite element methods. Second, based upon this theory we predict that switching the activation function to a piecewise linear B-spline, namely the Hat function, will remove this spectral bias, which we verify "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2208.04924","kind":"arxiv","version":3},"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/2208.04924/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T04:55:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"MhfrQlVUeAoS4nKHyCs58Mmye20+BeIk3lLCJErXpicT4Y2Uye9qkmfLwgQmi8TkvcyExS1dnb3ldREvT+gqCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T06:32:20.938925Z"},"content_sha256":"da0c13eec35c393ab470dfe4cfefb89aa41d871466329455c4366f7642677ad0","schema_version":"1.0","event_id":"sha256:da0c13eec35c393ab470dfe4cfefb89aa41d871466329455c4366f7642677ad0"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HE27HZUSH2OKMTKG5RQ43FF26T/bundle.json","state_url":"https://pith.science/pith/HE27HZUSH2OKMTKG5RQ43FF26T/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HE27HZUSH2OKMTKG5RQ43FF26T/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":"2026-08-09T06:32:20Z","links":{"resolver":"https://pith.science/pith/HE27HZUSH2OKMTKG5RQ43FF26T","bundle":"https://pith.science/pith/HE27HZUSH2OKMTKG5RQ43FF26T/bundle.json","state":"https://pith.science/pith/HE27HZUSH2OKMTKG5RQ43FF26T/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HE27HZUSH2OKMTKG5RQ43FF26T/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:HE27HZUSH2OKMTKG5RQ43FF26T","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":"cf38ddbf61c9000dd5c866ff06d0e60cc9af214df16c8ff3c6c4b92db79d70a0","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-08-09T17:40:57Z","title_canon_sha256":"97db8ce9aaa3e2cd73797e57d0a0ceaa685eeb4e16870ac46717cbff5ca72de4"},"schema_version":"1.0","source":{"id":"2208.04924","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2208.04924","created_at":"2026-07-05T04:55:06Z"},{"alias_kind":"arxiv_version","alias_value":"2208.04924v3","created_at":"2026-07-05T04:55:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2208.04924","created_at":"2026-07-05T04:55:06Z"},{"alias_kind":"pith_short_12","alias_value":"HE27HZUSH2OK","created_at":"2026-07-05T04:55:06Z"},{"alias_kind":"pith_short_16","alias_value":"HE27HZUSH2OKMTKG","created_at":"2026-07-05T04:55:06Z"},{"alias_kind":"pith_short_8","alias_value":"HE27HZUS","created_at":"2026-07-05T04:55:06Z"}],"graph_snapshots":[{"event_id":"sha256:da0c13eec35c393ab470dfe4cfefb89aa41d871466329455c4366f7642677ad0","target":"graph","created_at":"2026-07-05T04:55:06Z","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"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2208.04924/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Neural networks are universal function approximators which are known to generalize well despite being dramatically overparameterized. We study this phenomenon from the point of view of the spectral bias of neural networks. Our contributions are two-fold. First, we provide a theoretical explanation for the spectral bias of ReLU neural networks by leveraging connections with the theory of finite element methods. Second, based upon this theory we predict that switching the activation function to a piecewise linear B-spline, namely the Hat function, will remove this spectral bias, which we verify ","authors_text":"Jinchao Xu, Jonathan W. Siegel, Qingguo Hong, Qinyang Tan","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-08-09T17:40:57Z","title":"On the Activation Function Dependence of the Spectral Bias of Neural Networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2208.04924","kind":"arxiv","version":3},"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:fa281fc338baca358a9e2e4214d80e4da58d26af0faf826cdf4ed3f7a8ed1db2","target":"record","created_at":"2026-07-05T04:55:06Z","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":"cf38ddbf61c9000dd5c866ff06d0e60cc9af214df16c8ff3c6c4b92db79d70a0","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-08-09T17:40:57Z","title_canon_sha256":"97db8ce9aaa3e2cd73797e57d0a0ceaa685eeb4e16870ac46717cbff5ca72de4"},"schema_version":"1.0","source":{"id":"2208.04924","kind":"arxiv","version":3}},"canonical_sha256":"3935f3e6923e9ca64d46ec61cd94baf4c2c21f546c74618f0f193ac979ea9a49","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3935f3e6923e9ca64d46ec61cd94baf4c2c21f546c74618f0f193ac979ea9a49","first_computed_at":"2026-07-05T04:55:06.362107Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T04:55:06.362107Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"SIaLv4KrutJ64u/lPE2GdN5Xoef871dWDrdbYzXO/j69IqjZZCfdIx1eOqfTYNDFDLpxguNl3wh2Up+P2xG4Dg==","signature_status":"signed_v1","signed_at":"2026-07-05T04:55:06.362651Z","signed_message":"canonical_sha256_bytes"},"source_id":"2208.04924","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:fa281fc338baca358a9e2e4214d80e4da58d26af0faf826cdf4ed3f7a8ed1db2","sha256:da0c13eec35c393ab470dfe4cfefb89aa41d871466329455c4366f7642677ad0"],"state_sha256":"6ab51072f69a9ae3ee0f94ee5357620e0643e5b8eefa4c4db345ba29b381ee71"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xPe7IbDUnwlCyWhxxudEE3+oTGpGpcgTIiSO7jEMhD+lHNE/KjPL5pNE2eSj19/kEJMR3qheG8MeOFEPWT16CA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T06:32:20.943603Z","bundle_sha256":"425b3f4aabcb2e4d92205af41e1c63f343626001e74c5306f9cf7ccc48d33332"}}