{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:UURY23CBNTRAKIAC677PJD4YLS","short_pith_number":"pith:UURY23CB","schema_version":"1.0","canonical_sha256":"a5238d6c416ce2052002f7fef48f985ca430748e7a1c753a05711c056f82bef4","source":{"kind":"arxiv","id":"2508.19896","version":1},"attestation_state":"computed","paper":{"title":"NM-Hebb: Coupling Local Hebbian Plasticity with Metric Learning for More Accurate and Interpretable CNNs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Davorin Mili\\v{c}evi\\'c, Ratko Grbi\\'c","submitted_at":"2025-08-27T13:53:04Z","abstract_excerpt":"Deep Convolutional Neural Networks (CNNs) achieve high accuracy but often rely on purely global, gradient-based optimisation, which can lead to overfitting, redundant filters, and reduced interpretability. To address these limitations, we propose NM-Hebb, a two-phase training framework that integrates neuro-inspired local plasticity with distance-aware supervision. Phase 1 extends standard supervised training by jointly optimising a cross-entropy objective with two biologically inspired mechanisms: (i) a Hebbian regulariser that aligns the spatial mean of activations with the mean of the corre"},"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":"2508.19896","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-08-27T13:53:04Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"c31a924cac513811aa968cb9dd27a32fda35088636cf77be811d5467c7753a4d","abstract_canon_sha256":"db4abcd40a3ed1c23d631d4b4e17687f812dfe205c3b2f669d47dc1bfe52d70a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:00:19.607273Z","signature_b64":"E+AkM1JLZkRw7QJfF/9zwQdywtzUMv+86ha3iODuDrrd1XdMYtUQAVSP/4J8IpNe0NfP1/ot31nRbfq1ja0bBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a5238d6c416ce2052002f7fef48f985ca430748e7a1c753a05711c056f82bef4","last_reissued_at":"2026-07-05T12:00:19.606848Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:00:19.606848Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"NM-Hebb: Coupling Local Hebbian Plasticity with Metric Learning for More Accurate and Interpretable CNNs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Davorin Mili\\v{c}evi\\'c, Ratko Grbi\\'c","submitted_at":"2025-08-27T13:53:04Z","abstract_excerpt":"Deep Convolutional Neural Networks (CNNs) achieve high accuracy but often rely on purely global, gradient-based optimisation, which can lead to overfitting, redundant filters, and reduced interpretability. To address these limitations, we propose NM-Hebb, a two-phase training framework that integrates neuro-inspired local plasticity with distance-aware supervision. Phase 1 extends standard supervised training by jointly optimising a cross-entropy objective with two biologically inspired mechanisms: (i) a Hebbian regulariser that aligns the spatial mean of activations with the mean of the corre"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.19896","kind":"arxiv","version":1},"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/2508.19896/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":"2508.19896","created_at":"2026-07-05T12:00:19.606903+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.19896v1","created_at":"2026-07-05T12:00:19.606903+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.19896","created_at":"2026-07-05T12:00:19.606903+00:00"},{"alias_kind":"pith_short_12","alias_value":"UURY23CBNTRA","created_at":"2026-07-05T12:00:19.606903+00:00"},{"alias_kind":"pith_short_16","alias_value":"UURY23CBNTRAKIAC","created_at":"2026-07-05T12:00:19.606903+00:00"},{"alias_kind":"pith_short_8","alias_value":"UURY23CB","created_at":"2026-07-05T12:00:19.606903+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.16027","citing_title":"Constrained Hebbian Learning Supports Efficient Representational Allocation under Structural Constraints","ref_index":98,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UURY23CBNTRAKIAC677PJD4YLS","json":"https://pith.science/pith/UURY23CBNTRAKIAC677PJD4YLS.json","graph_json":"https://pith.science/api/pith-number/UURY23CBNTRAKIAC677PJD4YLS/graph.json","events_json":"https://pith.science/api/pith-number/UURY23CBNTRAKIAC677PJD4YLS/events.json","paper":"https://pith.science/paper/UURY23CB"},"agent_actions":{"view_html":"https://pith.science/pith/UURY23CBNTRAKIAC677PJD4YLS","download_json":"https://pith.science/pith/UURY23CBNTRAKIAC677PJD4YLS.json","view_paper":"https://pith.science/paper/UURY23CB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.19896&json=true","fetch_graph":"https://pith.science/api/pith-number/UURY23CBNTRAKIAC677PJD4YLS/graph.json","fetch_events":"https://pith.science/api/pith-number/UURY23CBNTRAKIAC677PJD4YLS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UURY23CBNTRAKIAC677PJD4YLS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UURY23CBNTRAKIAC677PJD4YLS/action/storage_attestation","attest_author":"https://pith.science/pith/UURY23CBNTRAKIAC677PJD4YLS/action/author_attestation","sign_citation":"https://pith.science/pith/UURY23CBNTRAKIAC677PJD4YLS/action/citation_signature","submit_replication":"https://pith.science/pith/UURY23CBNTRAKIAC677PJD4YLS/action/replication_record"}},"created_at":"2026-07-05T12:00:19.606903+00:00","updated_at":"2026-07-05T12:00:19.606903+00:00"}