{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:WWYJHAWETOG4CCEJOQVU3WM4AX","short_pith_number":"pith:WWYJHAWE","schema_version":"1.0","canonical_sha256":"b5b09382c49b8dc10889742b4dd99c05e61ed4dc23751c35761b413d9ef18e0d","source":{"kind":"arxiv","id":"2206.00823","version":3},"attestation_state":"computed","paper":{"title":"Beyond accuracy: generalization properties of bio-plausible temporal credit assignment rules","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","q-bio.NC"],"primary_cat":"cs.NE","authors_text":"Arna Ghosh, Blake A. Richards, Eric Shea-Brown, Guillaume Lajoie, Yuhan Helena Liu","submitted_at":"2022-06-02T01:39:08Z","abstract_excerpt":"To unveil how the brain learns, ongoing work seeks biologically-plausible approximations of gradient descent algorithms for training recurrent neural networks (RNNs). Yet, beyond task accuracy, it is unclear if such learning rules converge to solutions that exhibit different levels of generalization than their nonbiologically-plausible counterparts. Leveraging results from deep learning theory based on loss landscape curvature, we ask: how do biologically-plausible gradient approximations affect generalization? We first demonstrate that state-of-the-art biologically-plausible learning rules fo"},"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":"2206.00823","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.NE","submitted_at":"2022-06-02T01:39:08Z","cross_cats_sorted":["cs.AI","q-bio.NC"],"title_canon_sha256":"b3cbae4f2add48d96602e0a406321c38415af75841113b69a30df898566a88ef","abstract_canon_sha256":"cae1c327b39a475c50b67c0c42420829167e113fb9b267977f1f1f080f983e93"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:33:11.993200Z","signature_b64":"wieedGrWj11eGca+5slXeCb7KpceM4DbfipYFJPx2WX22nKfB8EdACcH/eHjTyFJnv0Gko2TwH3IGhYDF8UBAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b5b09382c49b8dc10889742b4dd99c05e61ed4dc23751c35761b413d9ef18e0d","last_reissued_at":"2026-07-05T05:33:11.992714Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:33:11.992714Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Beyond accuracy: generalization properties of bio-plausible temporal credit assignment rules","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","q-bio.NC"],"primary_cat":"cs.NE","authors_text":"Arna Ghosh, Blake A. Richards, Eric Shea-Brown, Guillaume Lajoie, Yuhan Helena Liu","submitted_at":"2022-06-02T01:39:08Z","abstract_excerpt":"To unveil how the brain learns, ongoing work seeks biologically-plausible approximations of gradient descent algorithms for training recurrent neural networks (RNNs). Yet, beyond task accuracy, it is unclear if such learning rules converge to solutions that exhibit different levels of generalization than their nonbiologically-plausible counterparts. Leveraging results from deep learning theory based on loss landscape curvature, we ask: how do biologically-plausible gradient approximations affect generalization? We first demonstrate that state-of-the-art biologically-plausible learning rules fo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2206.00823","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/2206.00823/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":"2206.00823","created_at":"2026-07-05T05:33:11.992773+00:00"},{"alias_kind":"arxiv_version","alias_value":"2206.00823v3","created_at":"2026-07-05T05:33:11.992773+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2206.00823","created_at":"2026-07-05T05:33:11.992773+00:00"},{"alias_kind":"pith_short_12","alias_value":"WWYJHAWETOG4","created_at":"2026-07-05T05:33:11.992773+00:00"},{"alias_kind":"pith_short_16","alias_value":"WWYJHAWETOG4CCEJ","created_at":"2026-07-05T05:33:11.992773+00:00"},{"alias_kind":"pith_short_8","alias_value":"WWYJHAWE","created_at":"2026-07-05T05:33:11.992773+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.06904","citing_title":"Can Biologically Plausible Temporal Credit Assignment Rules Match BPTT for Neural Similarity? E-prop as an Example","ref_index":13,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WWYJHAWETOG4CCEJOQVU3WM4AX","json":"https://pith.science/pith/WWYJHAWETOG4CCEJOQVU3WM4AX.json","graph_json":"https://pith.science/api/pith-number/WWYJHAWETOG4CCEJOQVU3WM4AX/graph.json","events_json":"https://pith.science/api/pith-number/WWYJHAWETOG4CCEJOQVU3WM4AX/events.json","paper":"https://pith.science/paper/WWYJHAWE"},"agent_actions":{"view_html":"https://pith.science/pith/WWYJHAWETOG4CCEJOQVU3WM4AX","download_json":"https://pith.science/pith/WWYJHAWETOG4CCEJOQVU3WM4AX.json","view_paper":"https://pith.science/paper/WWYJHAWE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2206.00823&json=true","fetch_graph":"https://pith.science/api/pith-number/WWYJHAWETOG4CCEJOQVU3WM4AX/graph.json","fetch_events":"https://pith.science/api/pith-number/WWYJHAWETOG4CCEJOQVU3WM4AX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WWYJHAWETOG4CCEJOQVU3WM4AX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WWYJHAWETOG4CCEJOQVU3WM4AX/action/storage_attestation","attest_author":"https://pith.science/pith/WWYJHAWETOG4CCEJOQVU3WM4AX/action/author_attestation","sign_citation":"https://pith.science/pith/WWYJHAWETOG4CCEJOQVU3WM4AX/action/citation_signature","submit_replication":"https://pith.science/pith/WWYJHAWETOG4CCEJOQVU3WM4AX/action/replication_record"}},"created_at":"2026-07-05T05:33:11.992773+00:00","updated_at":"2026-07-05T05:33:11.992773+00:00"}