{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:54JBPYOTNBDIYX6NTA3H3GCHCD","short_pith_number":"pith:54JBPYOT","schema_version":"1.0","canonical_sha256":"ef1217e1d368468c5fcd98367d984710cb673dd3d65d1f24fad6da3089a9850b","source":{"kind":"arxiv","id":"1904.07734","version":1},"attestation_state":"computed","paper":{"title":"Three scenarios for continual learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Andreas S. Tolias, Gido M. van de Ven","submitted_at":"2019-04-15T12:22:36Z","abstract_excerpt":"Standard artificial neural networks suffer from the well-known issue of catastrophic forgetting, making continual or lifelong learning difficult for machine learning. In recent years, numerous methods have been proposed for continual learning, but due to differences in evaluation protocols it is difficult to directly compare their performance. To enable more structured comparisons, we describe three continual learning scenarios based on whether at test time task identity is provided and--in case it is not--whether it must be inferred. Any sequence of well-defined tasks can be performed accordi"},"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":"1904.07734","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-04-15T12:22:36Z","cross_cats_sorted":["cs.AI","cs.CV","stat.ML"],"title_canon_sha256":"6cf049951fed9aa83005b7ae464938aa866f915149d0624606846f218d976cc0","abstract_canon_sha256":"ddd7b60af85ee92b5e5d4a57f553e7966c0e48b72fa4e105191e7c2f7bc45a6c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:48:19.523863Z","signature_b64":"q0W58mTyhkqxLGNl/CT51hkRvT4Ou7vlAtGkNsNNaf3f7BuO+N9dCVsw53xK3U0Rjy64ii1gjYQWq5evUIqUDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ef1217e1d368468c5fcd98367d984710cb673dd3d65d1f24fad6da3089a9850b","last_reissued_at":"2026-05-17T23:48:19.523217Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:48:19.523217Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Three scenarios for continual learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Andreas S. Tolias, Gido M. van de Ven","submitted_at":"2019-04-15T12:22:36Z","abstract_excerpt":"Standard artificial neural networks suffer from the well-known issue of catastrophic forgetting, making continual or lifelong learning difficult for machine learning. In recent years, numerous methods have been proposed for continual learning, but due to differences in evaluation protocols it is difficult to directly compare their performance. To enable more structured comparisons, we describe three continual learning scenarios based on whether at test time task identity is provided and--in case it is not--whether it must be inferred. Any sequence of well-defined tasks can be performed accordi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1904.07734","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":""},"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":"1904.07734","created_at":"2026-05-17T23:48:19.523322+00:00"},{"alias_kind":"arxiv_version","alias_value":"1904.07734v1","created_at":"2026-05-17T23:48:19.523322+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1904.07734","created_at":"2026-05-17T23:48:19.523322+00:00"},{"alias_kind":"pith_short_12","alias_value":"54JBPYOTNBDI","created_at":"2026-05-18T12:33:10.108867+00:00"},{"alias_kind":"pith_short_16","alias_value":"54JBPYOTNBDIYX6N","created_at":"2026-05-18T12:33:10.108867+00:00"},{"alias_kind":"pith_short_8","alias_value":"54JBPYOT","created_at":"2026-05-18T12:33:10.108867+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":18,"internal_anchor_count":9,"sample":[{"citing_arxiv_id":"2209.14774","citing_title":"RECALL: Rehearsal-free Continual Learning for Object Classification","ref_index":6,"is_internal_anchor":true},{"citing_arxiv_id":"2305.01507","citing_title":"A Parameter-free Adaptive Resonance Theory-based Topological Clustering Algorithm Capable of Continual Learning","ref_index":20,"is_internal_anchor":true},{"citing_arxiv_id":"2605.11617","citing_title":"MIST: Reliable Streaming Decision Trees for Online Class-Incremental Learning via McDiarmid Bound","ref_index":62,"is_internal_anchor":true},{"citing_arxiv_id":"2305.17493","citing_title":"The Curse of Recursion: Training on Generated Data Makes Models Forget","ref_index":10,"is_internal_anchor":true},{"citing_arxiv_id":"2505.18604","citing_title":"Exemplar-Free Continual Learning for State Space Models","ref_index":55,"is_internal_anchor":true},{"citing_arxiv_id":"2508.03941","citing_title":"Measuring the stability and plasticity of recommender systems","ref_index":12,"is_internal_anchor":true},{"citing_arxiv_id":"2602.07940","citing_title":"MePo: Meta Post-Refinement for Rehearsal-Free General Continual Learning","ref_index":6,"is_internal_anchor":true},{"citing_arxiv_id":"2601.16175","citing_title":"Learning to Discover at Test Time","ref_index":74,"is_internal_anchor":true},{"citing_arxiv_id":"2603.20410","citing_title":"SLE-FNO: Single-Layer Extensions for Task-Agnostic Continual Learning in Fourier Neural Operators","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2605.11617","citing_title":"MIST: Reliable Streaming Decision Trees for Online Class-Incremental Learning via McDiarmid Bound","ref_index":62,"is_internal_anchor":false},{"citing_arxiv_id":"2604.27031","citing_title":"NORACL: Neurogenesis for Oracle-free Resource-Adaptive Continual Learning","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10529","citing_title":"PrimeKG-CL: A Continual Graph Learning Benchmark on Evolving Biomedical Knowledge Graphs","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2604.24637","citing_title":"Cortex-Inspired Continual Learning: Unsupervised Instantiation and Recovery of Functional Task Networks","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05776","citing_title":"HEDP: A Hybrid Energy-Distance Prompt-based Framework for Domain Incremental Learning","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18270","citing_title":"Incremental learning for audio classification with Hebbian Deep Neural Networks","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07886","citing_title":"Characterizing and Correcting Effective Target Shift in Online Learning","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14259","citing_title":"Continual Learning for fMRI-Based Brain Disorder Diagnosis via Functional Connectivity Matrices Generative Replay","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2604.21927","citing_title":"Fine-Tuning Regimes Define Distinct Continual Learning Problems","ref_index":18,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/54JBPYOTNBDIYX6NTA3H3GCHCD","json":"https://pith.science/pith/54JBPYOTNBDIYX6NTA3H3GCHCD.json","graph_json":"https://pith.science/api/pith-number/54JBPYOTNBDIYX6NTA3H3GCHCD/graph.json","events_json":"https://pith.science/api/pith-number/54JBPYOTNBDIYX6NTA3H3GCHCD/events.json","paper":"https://pith.science/paper/54JBPYOT"},"agent_actions":{"view_html":"https://pith.science/pith/54JBPYOTNBDIYX6NTA3H3GCHCD","download_json":"https://pith.science/pith/54JBPYOTNBDIYX6NTA3H3GCHCD.json","view_paper":"https://pith.science/paper/54JBPYOT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1904.07734&json=true","fetch_graph":"https://pith.science/api/pith-number/54JBPYOTNBDIYX6NTA3H3GCHCD/graph.json","fetch_events":"https://pith.science/api/pith-number/54JBPYOTNBDIYX6NTA3H3GCHCD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/54JBPYOTNBDIYX6NTA3H3GCHCD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/54JBPYOTNBDIYX6NTA3H3GCHCD/action/storage_attestation","attest_author":"https://pith.science/pith/54JBPYOTNBDIYX6NTA3H3GCHCD/action/author_attestation","sign_citation":"https://pith.science/pith/54JBPYOTNBDIYX6NTA3H3GCHCD/action/citation_signature","submit_replication":"https://pith.science/pith/54JBPYOTNBDIYX6NTA3H3GCHCD/action/replication_record"}},"created_at":"2026-05-17T23:48:19.523322+00:00","updated_at":"2026-05-17T23:48:19.523322+00:00"}