{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:ET76PTNYSQJHSH2UTK7LFA4QH2","short_pith_number":"pith:ET76PTNY","schema_version":"1.0","canonical_sha256":"24ffe7cdb89412791f549abeb283903eafa3b83c7492a65afb99b385f8c9b300","source":{"kind":"arxiv","id":"2312.00246","version":4},"attestation_state":"computed","paper":{"title":"Directions of Curvature as an Explanation for Loss of Plasticity","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alex Lewandowski, Dale Schuurmans, Haruto Tanaka, Marlos C. Machado","submitted_at":"2023-11-30T23:24:45Z","abstract_excerpt":"Loss of plasticity is a phenomenon in which neural networks lose their ability to learn from new experience. Despite being empirically observed in several problem settings, little is understood about the mechanisms that lead to loss of plasticity. In this paper, we offer a consistent explanation for loss of plasticity: Neural networks lose directions of curvature during training and that loss of plasticity can be attributed to this reduction in curvature. To support such a claim, we provide a systematic investigation of loss of plasticity across continual learning tasks using MNIST, CIFAR-10 a"},"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":"2312.00246","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-11-30T23:24:45Z","cross_cats_sorted":[],"title_canon_sha256":"c6992487caf1f15ec2c9f5b3503fda14022e00caaca98db96ac1ec7520b13513","abstract_canon_sha256":"b054e311ed6de43ee382243903f3973ed56c0b09212f48cb1690e9e6f72bc80c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:16:10.445050Z","signature_b64":"F67zvU9xfu5bIMb1UOMPTtTX39XMkQKnv9FnOMWlW4+fPSsy4nNVBjClX+qxY420YXnjXNdQKYaU8wBSzMqMAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"24ffe7cdb89412791f549abeb283903eafa3b83c7492a65afb99b385f8c9b300","last_reissued_at":"2026-07-05T09:16:10.444579Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:16:10.444579Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Directions of Curvature as an Explanation for Loss of Plasticity","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alex Lewandowski, Dale Schuurmans, Haruto Tanaka, Marlos C. Machado","submitted_at":"2023-11-30T23:24:45Z","abstract_excerpt":"Loss of plasticity is a phenomenon in which neural networks lose their ability to learn from new experience. Despite being empirically observed in several problem settings, little is understood about the mechanisms that lead to loss of plasticity. In this paper, we offer a consistent explanation for loss of plasticity: Neural networks lose directions of curvature during training and that loss of plasticity can be attributed to this reduction in curvature. To support such a claim, we provide a systematic investigation of loss of plasticity across continual learning tasks using MNIST, CIFAR-10 a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.00246","kind":"arxiv","version":4},"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/2312.00246/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":"2312.00246","created_at":"2026-07-05T09:16:10.444637+00:00"},{"alias_kind":"arxiv_version","alias_value":"2312.00246v4","created_at":"2026-07-05T09:16:10.444637+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.00246","created_at":"2026-07-05T09:16:10.444637+00:00"},{"alias_kind":"pith_short_12","alias_value":"ET76PTNYSQJH","created_at":"2026-07-05T09:16:10.444637+00:00"},{"alias_kind":"pith_short_16","alias_value":"ET76PTNYSQJHSH2U","created_at":"2026-07-05T09:16:10.444637+00:00"},{"alias_kind":"pith_short_8","alias_value":"ET76PTNY","created_at":"2026-07-05T09:16:10.444637+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.28347","citing_title":"Agentic Safety is an Epistemic Property, Not a Behavioral One","ref_index":52,"is_internal_anchor":false},{"citing_arxiv_id":"2411.04832","citing_title":"Plasticity Loss in Deep Reinforcement Learning: A Survey","ref_index":63,"is_internal_anchor":false},{"citing_arxiv_id":"2509.22562","citing_title":"Activation Function Design Sustains Plasticity in Continual Learning","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09044","citing_title":"Predicting Plasticity in Deep Continual Learning: A Theoretical Perspective","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04712","citing_title":"SPHERE: Mitigating the Loss of Spectral Plasticity in Mixture-of-Experts for Deep Reinforcement Learning","ref_index":80,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06834","citing_title":"Attribution-Based Neuron Utility for Plasticity Restoration in Deep Networks","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04712","citing_title":"SPHERE: Mitigating the Loss of Spectral Plasticity in Mixture-of-Experts for Deep Reinforcement Learning","ref_index":80,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ET76PTNYSQJHSH2UTK7LFA4QH2","json":"https://pith.science/pith/ET76PTNYSQJHSH2UTK7LFA4QH2.json","graph_json":"https://pith.science/api/pith-number/ET76PTNYSQJHSH2UTK7LFA4QH2/graph.json","events_json":"https://pith.science/api/pith-number/ET76PTNYSQJHSH2UTK7LFA4QH2/events.json","paper":"https://pith.science/paper/ET76PTNY"},"agent_actions":{"view_html":"https://pith.science/pith/ET76PTNYSQJHSH2UTK7LFA4QH2","download_json":"https://pith.science/pith/ET76PTNYSQJHSH2UTK7LFA4QH2.json","view_paper":"https://pith.science/paper/ET76PTNY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2312.00246&json=true","fetch_graph":"https://pith.science/api/pith-number/ET76PTNYSQJHSH2UTK7LFA4QH2/graph.json","fetch_events":"https://pith.science/api/pith-number/ET76PTNYSQJHSH2UTK7LFA4QH2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ET76PTNYSQJHSH2UTK7LFA4QH2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ET76PTNYSQJHSH2UTK7LFA4QH2/action/storage_attestation","attest_author":"https://pith.science/pith/ET76PTNYSQJHSH2UTK7LFA4QH2/action/author_attestation","sign_citation":"https://pith.science/pith/ET76PTNYSQJHSH2UTK7LFA4QH2/action/citation_signature","submit_replication":"https://pith.science/pith/ET76PTNYSQJHSH2UTK7LFA4QH2/action/replication_record"}},"created_at":"2026-07-05T09:16:10.444637+00:00","updated_at":"2026-07-05T09:16:10.444637+00:00"}