{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:26SPF65SHLRM4J4TA7HHR4PEBX","short_pith_number":"pith:26SPF65S","schema_version":"1.0","canonical_sha256":"d7a4f2fbb23ae2ce279307ce78f1e40dd78a594b1d84c31e70eafe430b5a1bab","source":{"kind":"arxiv","id":"2305.09779","version":2},"attestation_state":"computed","paper":{"title":"A Scalable Walsh-Hadamard Regularizer to Overcome the Low-degree Spectral Bias of Neural Networks","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Ali Gorji, Andisheh Amrollahi, Andreas Krause","submitted_at":"2023-05-16T20:06:01Z","abstract_excerpt":"Despite the capacity of neural nets to learn arbitrary functions, models trained through gradient descent often exhibit a bias towards ``simpler'' functions. Various notions of simplicity have been introduced to characterize this behavior. Here, we focus on the case of neural networks with discrete (zero-one), high-dimensional, inputs through the lens of their Fourier (Walsh-Hadamard) transforms, where the notion of simplicity can be captured through the degree of the Fourier coefficients. We empirically show that neural networks have a tendency to learn lower-degree frequencies. We show how t"},"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":"2305.09779","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2023-05-16T20:06:01Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ad197ef23e77ef87d865488c685ba45349b49fb7fb4238131ab0f70e3e66fb7d","abstract_canon_sha256":"45968ae700d28bb6d4097a89611cea3926ae274fe8fd8bf1534e4a5769f69416"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:19:19.695831Z","signature_b64":"ECt2yLhlXcF1i6ZVr7UjDxOgnbx8n3snV55dALSE+7S9us5ePEbrf47VHGGNikN28qggY8c2qiivEoxMiNy7Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d7a4f2fbb23ae2ce279307ce78f1e40dd78a594b1d84c31e70eafe430b5a1bab","last_reissued_at":"2026-07-05T06:19:19.695380Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:19:19.695380Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Scalable Walsh-Hadamard Regularizer to Overcome the Low-degree Spectral Bias of Neural Networks","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Ali Gorji, Andisheh Amrollahi, Andreas Krause","submitted_at":"2023-05-16T20:06:01Z","abstract_excerpt":"Despite the capacity of neural nets to learn arbitrary functions, models trained through gradient descent often exhibit a bias towards ``simpler'' functions. Various notions of simplicity have been introduced to characterize this behavior. Here, we focus on the case of neural networks with discrete (zero-one), high-dimensional, inputs through the lens of their Fourier (Walsh-Hadamard) transforms, where the notion of simplicity can be captured through the degree of the Fourier coefficients. We empirically show that neural networks have a tendency to learn lower-degree frequencies. We show how t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.09779","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/2305.09779/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":"2305.09779","created_at":"2026-07-05T06:19:19.695437+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.09779v2","created_at":"2026-07-05T06:19:19.695437+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.09779","created_at":"2026-07-05T06:19:19.695437+00:00"},{"alias_kind":"pith_short_12","alias_value":"26SPF65SHLRM","created_at":"2026-07-05T06:19:19.695437+00:00"},{"alias_kind":"pith_short_16","alias_value":"26SPF65SHLRM4J4T","created_at":"2026-07-05T06:19:19.695437+00:00"},{"alias_kind":"pith_short_8","alias_value":"26SPF65S","created_at":"2026-07-05T06:19:19.695437+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/26SPF65SHLRM4J4TA7HHR4PEBX","json":"https://pith.science/pith/26SPF65SHLRM4J4TA7HHR4PEBX.json","graph_json":"https://pith.science/api/pith-number/26SPF65SHLRM4J4TA7HHR4PEBX/graph.json","events_json":"https://pith.science/api/pith-number/26SPF65SHLRM4J4TA7HHR4PEBX/events.json","paper":"https://pith.science/paper/26SPF65S"},"agent_actions":{"view_html":"https://pith.science/pith/26SPF65SHLRM4J4TA7HHR4PEBX","download_json":"https://pith.science/pith/26SPF65SHLRM4J4TA7HHR4PEBX.json","view_paper":"https://pith.science/paper/26SPF65S","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.09779&json=true","fetch_graph":"https://pith.science/api/pith-number/26SPF65SHLRM4J4TA7HHR4PEBX/graph.json","fetch_events":"https://pith.science/api/pith-number/26SPF65SHLRM4J4TA7HHR4PEBX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/26SPF65SHLRM4J4TA7HHR4PEBX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/26SPF65SHLRM4J4TA7HHR4PEBX/action/storage_attestation","attest_author":"https://pith.science/pith/26SPF65SHLRM4J4TA7HHR4PEBX/action/author_attestation","sign_citation":"https://pith.science/pith/26SPF65SHLRM4J4TA7HHR4PEBX/action/citation_signature","submit_replication":"https://pith.science/pith/26SPF65SHLRM4J4TA7HHR4PEBX/action/replication_record"}},"created_at":"2026-07-05T06:19:19.695437+00:00","updated_at":"2026-07-05T06:19:19.695437+00:00"}