{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:TBU5CBMHZ2HJUSQVXIDRGG6OEA","short_pith_number":"pith:TBU5CBMH","schema_version":"1.0","canonical_sha256":"9869d10587ce8e9a4a15ba07131bce20261a83e817dd327735023bdadfcf6833","source":{"kind":"arxiv","id":"2202.10203","version":1},"attestation_state":"computed","paper":{"title":"Learning Bayesian Sparse Networks with Full Experience Replay for Continual Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Anton van den Hengel, Dong Gong, Javen Qinfeng Shi, Qingsen Yan, Yuhang Liu","submitted_at":"2022-02-21T13:25:03Z","abstract_excerpt":"Continual Learning (CL) methods aim to enable machine learning models to learn new tasks without catastrophic forgetting of those that have been previously mastered. Existing CL approaches often keep a buffer of previously-seen samples, perform knowledge distillation, or use regularization techniques towards this goal. Despite their performance, they still suffer from interference across tasks which leads to catastrophic forgetting. To ameliorate this problem, we propose to only activate and select sparse neurons for learning current and past tasks at any stage. More parameters space and model"},"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":"2202.10203","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-02-21T13:25:03Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"3482335f2e8450310b86c48bdcd2b0af7408ed6051a8adfb3e9e75f81936e9fa","abstract_canon_sha256":"631d5fe2d20c9485acbe12a3fd6db22696541f6d03463eca788983de607b5227"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:58:39.845096Z","signature_b64":"Yhw+xqib82apxc74WvNlsDMomaramAAUXVx+R2gTDQZxQpQKjM5XLpk/agXqt29E06Qnq/6A81mWdfjxYVM9CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9869d10587ce8e9a4a15ba07131bce20261a83e817dd327735023bdadfcf6833","last_reissued_at":"2026-07-05T03:58:39.844654Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:58:39.844654Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning Bayesian Sparse Networks with Full Experience Replay for Continual Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Anton van den Hengel, Dong Gong, Javen Qinfeng Shi, Qingsen Yan, Yuhang Liu","submitted_at":"2022-02-21T13:25:03Z","abstract_excerpt":"Continual Learning (CL) methods aim to enable machine learning models to learn new tasks without catastrophic forgetting of those that have been previously mastered. Existing CL approaches often keep a buffer of previously-seen samples, perform knowledge distillation, or use regularization techniques towards this goal. Despite their performance, they still suffer from interference across tasks which leads to catastrophic forgetting. To ameliorate this problem, we propose to only activate and select sparse neurons for learning current and past tasks at any stage. More parameters space and model"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2202.10203","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/2202.10203/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":"2202.10203","created_at":"2026-07-05T03:58:39.844717+00:00"},{"alias_kind":"arxiv_version","alias_value":"2202.10203v1","created_at":"2026-07-05T03:58:39.844717+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2202.10203","created_at":"2026-07-05T03:58:39.844717+00:00"},{"alias_kind":"pith_short_12","alias_value":"TBU5CBMHZ2HJ","created_at":"2026-07-05T03:58:39.844717+00:00"},{"alias_kind":"pith_short_16","alias_value":"TBU5CBMHZ2HJUSQV","created_at":"2026-07-05T03:58:39.844717+00:00"},{"alias_kind":"pith_short_8","alias_value":"TBU5CBMH","created_at":"2026-07-05T03:58:39.844717+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.16884","citing_title":"The Importance of Being Lazy: Scaling Limits of Continual Learning","ref_index":2023,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TBU5CBMHZ2HJUSQVXIDRGG6OEA","json":"https://pith.science/pith/TBU5CBMHZ2HJUSQVXIDRGG6OEA.json","graph_json":"https://pith.science/api/pith-number/TBU5CBMHZ2HJUSQVXIDRGG6OEA/graph.json","events_json":"https://pith.science/api/pith-number/TBU5CBMHZ2HJUSQVXIDRGG6OEA/events.json","paper":"https://pith.science/paper/TBU5CBMH"},"agent_actions":{"view_html":"https://pith.science/pith/TBU5CBMHZ2HJUSQVXIDRGG6OEA","download_json":"https://pith.science/pith/TBU5CBMHZ2HJUSQVXIDRGG6OEA.json","view_paper":"https://pith.science/paper/TBU5CBMH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2202.10203&json=true","fetch_graph":"https://pith.science/api/pith-number/TBU5CBMHZ2HJUSQVXIDRGG6OEA/graph.json","fetch_events":"https://pith.science/api/pith-number/TBU5CBMHZ2HJUSQVXIDRGG6OEA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TBU5CBMHZ2HJUSQVXIDRGG6OEA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TBU5CBMHZ2HJUSQVXIDRGG6OEA/action/storage_attestation","attest_author":"https://pith.science/pith/TBU5CBMHZ2HJUSQVXIDRGG6OEA/action/author_attestation","sign_citation":"https://pith.science/pith/TBU5CBMHZ2HJUSQVXIDRGG6OEA/action/citation_signature","submit_replication":"https://pith.science/pith/TBU5CBMHZ2HJUSQVXIDRGG6OEA/action/replication_record"}},"created_at":"2026-07-05T03:58:39.844717+00:00","updated_at":"2026-07-05T03:58:39.844717+00:00"}