{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:FT4KAS4KAZNZRD4EKBC3ZAV45N","short_pith_number":"pith:FT4KAS4K","schema_version":"1.0","canonical_sha256":"2cf8a04b8a065b988f845045bc82bceb54f86b9607e07e9b3980724da2b15dd6","source":{"kind":"arxiv","id":"1804.00379","version":2},"attestation_state":"computed","paper":{"title":"Recall Traces: Backtracking Models for Efficient Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Anirudh Goyal, Hugo Larochelle, Philemon Brakel, Sergey Levine, Soumye Singhal, Timothy Lillicrap, William Fedus, Yoshua Bengio","submitted_at":"2018-04-02T03:02:33Z","abstract_excerpt":"In many environments only a tiny subset of all states yield high reward. In these cases, few of the interactions with the environment provide a relevant learning signal. Hence, we may want to preferentially train on those high-reward states and the probable trajectories leading to them. To this end, we advocate for the use of a backtracking model that predicts the preceding states that terminate at a given high-reward state. We can train a model which, starting from a high value state (or one that is estimated to have high value), predicts and sample for which the (state, action)-tuples may ha"},"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":"1804.00379","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-04-02T03:02:33Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"2e225e4d1e70064820a4cf744612e47e87f1e6cef6d90a5e58c2160ffb23234a","abstract_canon_sha256":"57dd3d8e144bf889cd5b24fdee936bd874b7887cabd8864d236dfe71139c28b6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:55:21.068950Z","signature_b64":"Xa4tbO8NnJFwWDpfYgcIgwOHNmTgte+Ynvj5KuNxk47rBmG9lkSYZtJrJp7COyLzMTbxED5bGMhtYzvWjithCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2cf8a04b8a065b988f845045bc82bceb54f86b9607e07e9b3980724da2b15dd6","last_reissued_at":"2026-05-17T23:55:21.068515Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:55:21.068515Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Recall Traces: Backtracking Models for Efficient Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Anirudh Goyal, Hugo Larochelle, Philemon Brakel, Sergey Levine, Soumye Singhal, Timothy Lillicrap, William Fedus, Yoshua Bengio","submitted_at":"2018-04-02T03:02:33Z","abstract_excerpt":"In many environments only a tiny subset of all states yield high reward. In these cases, few of the interactions with the environment provide a relevant learning signal. Hence, we may want to preferentially train on those high-reward states and the probable trajectories leading to them. To this end, we advocate for the use of a backtracking model that predicts the preceding states that terminate at a given high-reward state. We can train a model which, starting from a high value state (or one that is estimated to have high value), predicts and sample for which the (state, action)-tuples may ha"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1804.00379","kind":"arxiv","version":2},"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":"1804.00379","created_at":"2026-05-17T23:55:21.068590+00:00"},{"alias_kind":"arxiv_version","alias_value":"1804.00379v2","created_at":"2026-05-17T23:55:21.068590+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1804.00379","created_at":"2026-05-17T23:55:21.068590+00:00"},{"alias_kind":"pith_short_12","alias_value":"FT4KAS4KAZNZ","created_at":"2026-05-18T12:32:25.280505+00:00"},{"alias_kind":"pith_short_16","alias_value":"FT4KAS4KAZNZRD4E","created_at":"2026-05-18T12:32:25.280505+00:00"},{"alias_kind":"pith_short_8","alias_value":"FT4KAS4K","created_at":"2026-05-18T12:32:25.280505+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"1907.01285","citing_title":"Learning the Arrow of Time","ref_index":6,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FT4KAS4KAZNZRD4EKBC3ZAV45N","json":"https://pith.science/pith/FT4KAS4KAZNZRD4EKBC3ZAV45N.json","graph_json":"https://pith.science/api/pith-number/FT4KAS4KAZNZRD4EKBC3ZAV45N/graph.json","events_json":"https://pith.science/api/pith-number/FT4KAS4KAZNZRD4EKBC3ZAV45N/events.json","paper":"https://pith.science/paper/FT4KAS4K"},"agent_actions":{"view_html":"https://pith.science/pith/FT4KAS4KAZNZRD4EKBC3ZAV45N","download_json":"https://pith.science/pith/FT4KAS4KAZNZRD4EKBC3ZAV45N.json","view_paper":"https://pith.science/paper/FT4KAS4K","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1804.00379&json=true","fetch_graph":"https://pith.science/api/pith-number/FT4KAS4KAZNZRD4EKBC3ZAV45N/graph.json","fetch_events":"https://pith.science/api/pith-number/FT4KAS4KAZNZRD4EKBC3ZAV45N/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FT4KAS4KAZNZRD4EKBC3ZAV45N/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FT4KAS4KAZNZRD4EKBC3ZAV45N/action/storage_attestation","attest_author":"https://pith.science/pith/FT4KAS4KAZNZRD4EKBC3ZAV45N/action/author_attestation","sign_citation":"https://pith.science/pith/FT4KAS4KAZNZRD4EKBC3ZAV45N/action/citation_signature","submit_replication":"https://pith.science/pith/FT4KAS4KAZNZRD4EKBC3ZAV45N/action/replication_record"}},"created_at":"2026-05-17T23:55:21.068590+00:00","updated_at":"2026-05-17T23:55:21.068590+00:00"}