{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:WPKXXWW5CFIAMFVYWQWYTGIJ3Y","short_pith_number":"pith:WPKXXWW5","schema_version":"1.0","canonical_sha256":"b3d57bdadd11500616b8b42d899909de052cef46b6ee6810c27a3fdab0925ce7","source":{"kind":"arxiv","id":"1909.10008","version":1},"attestation_state":"computed","paper":{"title":"Multi-task Learning and Catastrophic Forgetting in Continual Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Francisco S. Melo, Jo\\~ao Dias, Jo\\~ao Ribeiro","submitted_at":"2019-09-22T14:00:29Z","abstract_excerpt":"In this paper we investigate two hypothesis regarding the use of deep reinforcement learning in multiple tasks. The first hypothesis is driven by the question of whether a deep reinforcement learning algorithm, trained on two similar tasks, is able to outperform two single-task, individually trained algorithms, by more efficiently learning a new, similar task, that none of the three algorithms has encountered before. The second hypothesis is driven by the question of whether the same multi-task deep RL algorithm, trained on two similar tasks and augmented with elastic weight consolidation (EWC"},"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":"1909.10008","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-09-22T14:00:29Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"e148522a2527b52aa1705a0ef218df0a3462f161ebd1e4b2baa5d3cd520c02d2","abstract_canon_sha256":"bdac23a67d2b36e826660ff420dc34b8801400ff726239eefa3b029606da1bec"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:06:16.357210Z","signature_b64":"vYVuUimx92CW9uTfDglnGScLcFI+PZJDBVZHRBXBv9MLmS0AleC8wTUxF73I8bGfw707uM85qsm0fLbc4c7TAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b3d57bdadd11500616b8b42d899909de052cef46b6ee6810c27a3fdab0925ce7","last_reissued_at":"2026-07-05T00:06:16.356736Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:06:16.356736Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multi-task Learning and Catastrophic Forgetting in Continual Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Francisco S. Melo, Jo\\~ao Dias, Jo\\~ao Ribeiro","submitted_at":"2019-09-22T14:00:29Z","abstract_excerpt":"In this paper we investigate two hypothesis regarding the use of deep reinforcement learning in multiple tasks. The first hypothesis is driven by the question of whether a deep reinforcement learning algorithm, trained on two similar tasks, is able to outperform two single-task, individually trained algorithms, by more efficiently learning a new, similar task, that none of the three algorithms has encountered before. The second hypothesis is driven by the question of whether the same multi-task deep RL algorithm, trained on two similar tasks and augmented with elastic weight consolidation (EWC"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1909.10008","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/1909.10008/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":"1909.10008","created_at":"2026-07-05T00:06:16.356789+00:00"},{"alias_kind":"arxiv_version","alias_value":"1909.10008v1","created_at":"2026-07-05T00:06:16.356789+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1909.10008","created_at":"2026-07-05T00:06:16.356789+00:00"},{"alias_kind":"pith_short_12","alias_value":"WPKXXWW5CFIA","created_at":"2026-07-05T00:06:16.356789+00:00"},{"alias_kind":"pith_short_16","alias_value":"WPKXXWW5CFIAMFVY","created_at":"2026-07-05T00:06:16.356789+00:00"},{"alias_kind":"pith_short_8","alias_value":"WPKXXWW5","created_at":"2026-07-05T00:06:16.356789+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.12029","citing_title":"Growable and Interpretable Neural Control with Online Continual Learning for Autonomous Lifelong Locomotion Learning Machines","ref_index":45,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WPKXXWW5CFIAMFVYWQWYTGIJ3Y","json":"https://pith.science/pith/WPKXXWW5CFIAMFVYWQWYTGIJ3Y.json","graph_json":"https://pith.science/api/pith-number/WPKXXWW5CFIAMFVYWQWYTGIJ3Y/graph.json","events_json":"https://pith.science/api/pith-number/WPKXXWW5CFIAMFVYWQWYTGIJ3Y/events.json","paper":"https://pith.science/paper/WPKXXWW5"},"agent_actions":{"view_html":"https://pith.science/pith/WPKXXWW5CFIAMFVYWQWYTGIJ3Y","download_json":"https://pith.science/pith/WPKXXWW5CFIAMFVYWQWYTGIJ3Y.json","view_paper":"https://pith.science/paper/WPKXXWW5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1909.10008&json=true","fetch_graph":"https://pith.science/api/pith-number/WPKXXWW5CFIAMFVYWQWYTGIJ3Y/graph.json","fetch_events":"https://pith.science/api/pith-number/WPKXXWW5CFIAMFVYWQWYTGIJ3Y/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WPKXXWW5CFIAMFVYWQWYTGIJ3Y/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WPKXXWW5CFIAMFVYWQWYTGIJ3Y/action/storage_attestation","attest_author":"https://pith.science/pith/WPKXXWW5CFIAMFVYWQWYTGIJ3Y/action/author_attestation","sign_citation":"https://pith.science/pith/WPKXXWW5CFIAMFVYWQWYTGIJ3Y/action/citation_signature","submit_replication":"https://pith.science/pith/WPKXXWW5CFIAMFVYWQWYTGIJ3Y/action/replication_record"}},"created_at":"2026-07-05T00:06:16.356789+00:00","updated_at":"2026-07-05T00:06:16.356789+00:00"}