{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:IBXMEMXBC5P6SLFVOSOYNWEG2Y","short_pith_number":"pith:IBXMEMXB","schema_version":"1.0","canonical_sha256":"406ec232e1175fe92cb5749d86d886d63ac1746dda92df7bf9d3da590c7ce114","source":{"kind":"arxiv","id":"2310.00965","version":8},"attestation_state":"computed","paper":{"title":"Node Perturbation Can Effectively Train Multi-Layer Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Marcel van Gerven, Nasir Ahmad, Sander Dalm","submitted_at":"2023-10-02T08:12:51Z","abstract_excerpt":"Backpropagation (BP) remains the dominant and most successful method for training parameters of deep neural network models. However, BP relies on two computationally distinct phases, does not provide a satisfactory explanation of biological learning, and can be challenging to apply for training of networks with discontinuities or noisy node dynamics. By comparison, node perturbation (NP), also known as activity-perturbed forward gradients, proposes learning by the injection of noise into network activations, and subsequent measurement of the induced loss change. NP relies on two forward (infer"},"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":"2310.00965","kind":"arxiv","version":8},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-10-02T08:12:51Z","cross_cats_sorted":[],"title_canon_sha256":"c8e8037545c61ec7863003f6ef767ae36c204100faa6adc17d881cffdb5c21a3","abstract_canon_sha256":"246df41eaa3678088ac3d0968028574b30a9626ddaa400c22c6b88c16438e142"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-03T01:05:42.444580Z","signature_b64":"NySzXSCuKzrf4SNdXsxHvgEEscaA/kNS+qdlYuDIUhMqK7WCJDReuByGs/riZprP0zCHTfXtAl0a7VXjX7IyCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"406ec232e1175fe92cb5749d86d886d63ac1746dda92df7bf9d3da590c7ce114","last_reissued_at":"2026-06-03T01:05:42.444180Z","signature_status":"signed_v1","first_computed_at":"2026-06-03T01:05:42.444180Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Node Perturbation Can Effectively Train Multi-Layer Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Marcel van Gerven, Nasir Ahmad, Sander Dalm","submitted_at":"2023-10-02T08:12:51Z","abstract_excerpt":"Backpropagation (BP) remains the dominant and most successful method for training parameters of deep neural network models. However, BP relies on two computationally distinct phases, does not provide a satisfactory explanation of biological learning, and can be challenging to apply for training of networks with discontinuities or noisy node dynamics. By comparison, node perturbation (NP), also known as activity-perturbed forward gradients, proposes learning by the injection of noise into network activations, and subsequent measurement of the induced loss change. NP relies on two forward (infer"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.00965","kind":"arxiv","version":8},"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/2310.00965/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":"2310.00965","created_at":"2026-06-03T01:05:42.444235+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.00965v8","created_at":"2026-06-03T01:05:42.444235+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.00965","created_at":"2026-06-03T01:05:42.444235+00:00"},{"alias_kind":"pith_short_12","alias_value":"IBXMEMXBC5P6","created_at":"2026-06-03T01:05:42.444235+00:00"},{"alias_kind":"pith_short_16","alias_value":"IBXMEMXBC5P6SLFV","created_at":"2026-06-03T01:05:42.444235+00:00"},{"alias_kind":"pith_short_8","alias_value":"IBXMEMXB","created_at":"2026-06-03T01:05:42.444235+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.01928","citing_title":"Training Non-Differentiable Networks via Optimal Transport","ref_index":51,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IBXMEMXBC5P6SLFVOSOYNWEG2Y","json":"https://pith.science/pith/IBXMEMXBC5P6SLFVOSOYNWEG2Y.json","graph_json":"https://pith.science/api/pith-number/IBXMEMXBC5P6SLFVOSOYNWEG2Y/graph.json","events_json":"https://pith.science/api/pith-number/IBXMEMXBC5P6SLFVOSOYNWEG2Y/events.json","paper":"https://pith.science/paper/IBXMEMXB"},"agent_actions":{"view_html":"https://pith.science/pith/IBXMEMXBC5P6SLFVOSOYNWEG2Y","download_json":"https://pith.science/pith/IBXMEMXBC5P6SLFVOSOYNWEG2Y.json","view_paper":"https://pith.science/paper/IBXMEMXB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.00965&json=true","fetch_graph":"https://pith.science/api/pith-number/IBXMEMXBC5P6SLFVOSOYNWEG2Y/graph.json","fetch_events":"https://pith.science/api/pith-number/IBXMEMXBC5P6SLFVOSOYNWEG2Y/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IBXMEMXBC5P6SLFVOSOYNWEG2Y/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IBXMEMXBC5P6SLFVOSOYNWEG2Y/action/storage_attestation","attest_author":"https://pith.science/pith/IBXMEMXBC5P6SLFVOSOYNWEG2Y/action/author_attestation","sign_citation":"https://pith.science/pith/IBXMEMXBC5P6SLFVOSOYNWEG2Y/action/citation_signature","submit_replication":"https://pith.science/pith/IBXMEMXBC5P6SLFVOSOYNWEG2Y/action/replication_record"}},"created_at":"2026-06-03T01:05:42.444235+00:00","updated_at":"2026-06-03T01:05:42.444235+00:00"}