{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:3AFEC74CCK73NJRZLJDHDC3HDV","short_pith_number":"pith:3AFEC74C","schema_version":"1.0","canonical_sha256":"d80a417f8212bfb6a6395a46718b671d4ab777f439207de973ed7d0073f3288e","source":{"kind":"arxiv","id":"2006.10974","version":3},"attestation_state":"computed","paper":{"title":"Optimization and Generalization of Regularization-Based Continual Learning: a Loss Approximation Viewpoint","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Alex Mott, Ang Li, Dong Yin, Mehrdad Farajtabar, Nir Levine","submitted_at":"2020-06-19T06:08:40Z","abstract_excerpt":"Neural networks have achieved remarkable success in many cognitive tasks. However, when they are trained sequentially on multiple tasks without access to old data, their performance on early tasks tend to drop significantly. This problem is often referred to as catastrophic forgetting, a key challenge in continual learning of neural networks. The regularization-based approach is one of the primary classes of methods to alleviate catastrophic forgetting. In this paper, we provide a novel viewpoint of regularization-based continual learning by formulating it as a second-order Taylor approximatio"},"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":"2006.10974","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-06-19T06:08:40Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"be2ee54a51b7e823d37ba70ba21b68eac032496a6805db6490056e9200ebefb2","abstract_canon_sha256":"f235fe6fe6d7ddf81339349fb6249c4976555a44cddf3eb6fec7262e455e78cd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:13:46.287799Z","signature_b64":"oq5rNSleJTJz+Aa2lG42QOpOVjS4JRiVcnpn+gzc/tJNw9XGFDOS78KITr+PLSHS9065ac0Po2hfGp8ca0zKDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d80a417f8212bfb6a6395a46718b671d4ab777f439207de973ed7d0073f3288e","last_reissued_at":"2026-07-05T02:13:46.287378Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:13:46.287378Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Optimization and Generalization of Regularization-Based Continual Learning: a Loss Approximation Viewpoint","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Alex Mott, Ang Li, Dong Yin, Mehrdad Farajtabar, Nir Levine","submitted_at":"2020-06-19T06:08:40Z","abstract_excerpt":"Neural networks have achieved remarkable success in many cognitive tasks. However, when they are trained sequentially on multiple tasks without access to old data, their performance on early tasks tend to drop significantly. This problem is often referred to as catastrophic forgetting, a key challenge in continual learning of neural networks. The regularization-based approach is one of the primary classes of methods to alleviate catastrophic forgetting. In this paper, we provide a novel viewpoint of regularization-based continual learning by formulating it as a second-order Taylor approximatio"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2006.10974","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/2006.10974/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":"2006.10974","created_at":"2026-07-05T02:13:46.287438+00:00"},{"alias_kind":"arxiv_version","alias_value":"2006.10974v3","created_at":"2026-07-05T02:13:46.287438+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2006.10974","created_at":"2026-07-05T02:13:46.287438+00:00"},{"alias_kind":"pith_short_12","alias_value":"3AFEC74CCK73","created_at":"2026-07-05T02:13:46.287438+00:00"},{"alias_kind":"pith_short_16","alias_value":"3AFEC74CCK73NJRZ","created_at":"2026-07-05T02:13:46.287438+00:00"},{"alias_kind":"pith_short_8","alias_value":"3AFEC74C","created_at":"2026-07-05T02:13:46.287438+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.05609","citing_title":"To Retain or to Adapt? Generalizing Continual Learning","ref_index":58,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3AFEC74CCK73NJRZLJDHDC3HDV","json":"https://pith.science/pith/3AFEC74CCK73NJRZLJDHDC3HDV.json","graph_json":"https://pith.science/api/pith-number/3AFEC74CCK73NJRZLJDHDC3HDV/graph.json","events_json":"https://pith.science/api/pith-number/3AFEC74CCK73NJRZLJDHDC3HDV/events.json","paper":"https://pith.science/paper/3AFEC74C"},"agent_actions":{"view_html":"https://pith.science/pith/3AFEC74CCK73NJRZLJDHDC3HDV","download_json":"https://pith.science/pith/3AFEC74CCK73NJRZLJDHDC3HDV.json","view_paper":"https://pith.science/paper/3AFEC74C","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2006.10974&json=true","fetch_graph":"https://pith.science/api/pith-number/3AFEC74CCK73NJRZLJDHDC3HDV/graph.json","fetch_events":"https://pith.science/api/pith-number/3AFEC74CCK73NJRZLJDHDC3HDV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3AFEC74CCK73NJRZLJDHDC3HDV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3AFEC74CCK73NJRZLJDHDC3HDV/action/storage_attestation","attest_author":"https://pith.science/pith/3AFEC74CCK73NJRZLJDHDC3HDV/action/author_attestation","sign_citation":"https://pith.science/pith/3AFEC74CCK73NJRZLJDHDC3HDV/action/citation_signature","submit_replication":"https://pith.science/pith/3AFEC74CCK73NJRZLJDHDC3HDV/action/replication_record"}},"created_at":"2026-07-05T02:13:46.287438+00:00","updated_at":"2026-07-05T02:13:46.287438+00:00"}