{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:IATTN4REXT4GBU4BYFHGNOGDVO","short_pith_number":"pith:IATTN4RE","schema_version":"1.0","canonical_sha256":"402736f224bcf860d381c14e66b8c3abb34e929f3c42589131cd6984365523c3","source":{"kind":"arxiv","id":"1909.01387","version":1},"attestation_state":"computed","paper":{"title":"Making Efficient Use of Demonstrations to Solve Hard Exploration Problems","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Bobak Shahriari, Caglar Gulcehre, Duncan Williams, Gabriel Barth-Maron, Hubert Soyer, Matt Hoffman, Misha Denil, Nando de Freitas, Neil Rabinowitz, Richard Tanburn, Steven Kapturowski, Tom Le Paine, Worlds Team, Ziyu Wang","submitted_at":"2019-09-03T18:20:48Z","abstract_excerpt":"This paper introduces R2D3, an agent that makes efficient use of demonstrations to solve hard exploration problems in partially observable environments with highly variable initial conditions. We also introduce a suite of eight tasks that combine these three properties, and show that R2D3 can solve several of the tasks where other state of the art methods (both with and without demonstrations) fail to see even a single successful trajectory after tens of billions of steps of exploration."},"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":"1909.01387","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-09-03T18:20:48Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b388e9874d112092568ec6f2a7fc2dbf7d9bb10506ea9b0f8cee2f441dc11a20","abstract_canon_sha256":"9cb0a79cef24d4680b0d12d55878f7307e01c482615e4357fe437c8723b467c4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:02:15.681068Z","signature_b64":"QOtkP9WxZgmC92vx1gd1+pM4G1DXrXfX8kkZTRpmw8UpuKxU+oLLbrWOBa4U5vFuwrsEMokudxE5UxxriqC/Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"402736f224bcf860d381c14e66b8c3abb34e929f3c42589131cd6984365523c3","last_reissued_at":"2026-07-05T00:02:15.680592Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:02:15.680592Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Making Efficient Use of Demonstrations to Solve Hard Exploration Problems","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Bobak Shahriari, Caglar Gulcehre, Duncan Williams, Gabriel Barth-Maron, Hubert Soyer, Matt Hoffman, Misha Denil, Nando de Freitas, Neil Rabinowitz, Richard Tanburn, Steven Kapturowski, Tom Le Paine, Worlds Team, Ziyu Wang","submitted_at":"2019-09-03T18:20:48Z","abstract_excerpt":"This paper introduces R2D3, an agent that makes efficient use of demonstrations to solve hard exploration problems in partially observable environments with highly variable initial conditions. We also introduce a suite of eight tasks that combine these three properties, and show that R2D3 can solve several of the tasks where other state of the art methods (both with and without demonstrations) fail to see even a single successful trajectory after tens of billions of steps of exploration."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1909.01387","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/1909.01387/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":"1909.01387","created_at":"2026-07-05T00:02:15.680654+00:00"},{"alias_kind":"arxiv_version","alias_value":"1909.01387v1","created_at":"2026-07-05T00:02:15.680654+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1909.01387","created_at":"2026-07-05T00:02:15.680654+00:00"},{"alias_kind":"pith_short_12","alias_value":"IATTN4REXT4G","created_at":"2026-07-05T00:02:15.680654+00:00"},{"alias_kind":"pith_short_16","alias_value":"IATTN4REXT4GBU4B","created_at":"2026-07-05T00:02:15.680654+00:00"},{"alias_kind":"pith_short_8","alias_value":"IATTN4RE","created_at":"2026-07-05T00:02:15.680654+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.19592","citing_title":"Implicit Action Chunking for Smooth Continuous Control","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2507.14995","citing_title":"LLM-Enhanced Multi-Agent Reinforcement Learning with Expert Workflow for Real-Time P2P Energy Trading","ref_index":6,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IATTN4REXT4GBU4BYFHGNOGDVO","json":"https://pith.science/pith/IATTN4REXT4GBU4BYFHGNOGDVO.json","graph_json":"https://pith.science/api/pith-number/IATTN4REXT4GBU4BYFHGNOGDVO/graph.json","events_json":"https://pith.science/api/pith-number/IATTN4REXT4GBU4BYFHGNOGDVO/events.json","paper":"https://pith.science/paper/IATTN4RE"},"agent_actions":{"view_html":"https://pith.science/pith/IATTN4REXT4GBU4BYFHGNOGDVO","download_json":"https://pith.science/pith/IATTN4REXT4GBU4BYFHGNOGDVO.json","view_paper":"https://pith.science/paper/IATTN4RE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1909.01387&json=true","fetch_graph":"https://pith.science/api/pith-number/IATTN4REXT4GBU4BYFHGNOGDVO/graph.json","fetch_events":"https://pith.science/api/pith-number/IATTN4REXT4GBU4BYFHGNOGDVO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IATTN4REXT4GBU4BYFHGNOGDVO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IATTN4REXT4GBU4BYFHGNOGDVO/action/storage_attestation","attest_author":"https://pith.science/pith/IATTN4REXT4GBU4BYFHGNOGDVO/action/author_attestation","sign_citation":"https://pith.science/pith/IATTN4REXT4GBU4BYFHGNOGDVO/action/citation_signature","submit_replication":"https://pith.science/pith/IATTN4REXT4GBU4BYFHGNOGDVO/action/replication_record"}},"created_at":"2026-07-05T00:02:15.680654+00:00","updated_at":"2026-07-05T00:02:15.680654+00:00"}