{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:DUCNE4AMJH2QPECADCQQBUC262","short_pith_number":"pith:DUCNE4AM","schema_version":"1.0","canonical_sha256":"1d04d2700c49f507904018a100d05af6bb91df136dee96e2ed18811f367c9255","source":{"kind":"arxiv","id":"1709.10089","version":2},"attestation_state":"computed","paper":{"title":"Overcoming Exploration in Reinforcement Learning with Demonstrations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.NE","cs.RO"],"primary_cat":"cs.LG","authors_text":"Ashvin Nair, Bob McGrew, Marcin Andrychowicz, Pieter Abbeel, Wojciech Zaremba","submitted_at":"2017-09-28T17:51:48Z","abstract_excerpt":"Exploration in environments with sparse rewards has been a persistent problem in reinforcement learning (RL). Many tasks are natural to specify with a sparse reward, and manually shaping a reward function can result in suboptimal performance. However, finding a non-zero reward is exponentially more difficult with increasing task horizon or action dimensionality. This puts many real-world tasks out of practical reach of RL methods. In this work, we use demonstrations to overcome the exploration problem and successfully learn to perform long-horizon, multi-step robotics tasks with continuous con"},"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":"1709.10089","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-09-28T17:51:48Z","cross_cats_sorted":["cs.AI","cs.NE","cs.RO"],"title_canon_sha256":"b82cb3dcfe88db3f6f906ad539222ab5a63bc94acd5fba6f2c79c62969841369","abstract_canon_sha256":"4aed394a5018bd7a864ec59793c704619bc48062ece3e1734db28e1ccfb29cb3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:22:39.464934Z","signature_b64":"OYSfssTGQfbmr9mYkps5aAk4o1AiKXaoMpyKobuGQJWOiB3imT2z1t9wbd/YNJoLncgPQOWMlM0bMbPtL3gaBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1d04d2700c49f507904018a100d05af6bb91df136dee96e2ed18811f367c9255","last_reissued_at":"2026-05-18T00:22:39.464291Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:22:39.464291Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Overcoming Exploration in Reinforcement Learning with Demonstrations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.NE","cs.RO"],"primary_cat":"cs.LG","authors_text":"Ashvin Nair, Bob McGrew, Marcin Andrychowicz, Pieter Abbeel, Wojciech Zaremba","submitted_at":"2017-09-28T17:51:48Z","abstract_excerpt":"Exploration in environments with sparse rewards has been a persistent problem in reinforcement learning (RL). Many tasks are natural to specify with a sparse reward, and manually shaping a reward function can result in suboptimal performance. However, finding a non-zero reward is exponentially more difficult with increasing task horizon or action dimensionality. This puts many real-world tasks out of practical reach of RL methods. In this work, we use demonstrations to overcome the exploration problem and successfully learn to perform long-horizon, multi-step robotics tasks with continuous con"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1709.10089","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":"1709.10089","created_at":"2026-05-18T00:22:39.464401+00:00"},{"alias_kind":"arxiv_version","alias_value":"1709.10089v2","created_at":"2026-05-18T00:22:39.464401+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1709.10089","created_at":"2026-05-18T00:22:39.464401+00:00"},{"alias_kind":"pith_short_12","alias_value":"DUCNE4AMJH2Q","created_at":"2026-05-18T12:31:12.930513+00:00"},{"alias_kind":"pith_short_16","alias_value":"DUCNE4AMJH2QPECA","created_at":"2026-05-18T12:31:12.930513+00:00"},{"alias_kind":"pith_short_8","alias_value":"DUCNE4AM","created_at":"2026-05-18T12:31:12.930513+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.07986","citing_title":"EXPO: Stable Reinforcement Learning with Expressive Policies","ref_index":18,"is_internal_anchor":true},{"citing_arxiv_id":"2308.00352","citing_title":"MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework","ref_index":264,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DUCNE4AMJH2QPECADCQQBUC262","json":"https://pith.science/pith/DUCNE4AMJH2QPECADCQQBUC262.json","graph_json":"https://pith.science/api/pith-number/DUCNE4AMJH2QPECADCQQBUC262/graph.json","events_json":"https://pith.science/api/pith-number/DUCNE4AMJH2QPECADCQQBUC262/events.json","paper":"https://pith.science/paper/DUCNE4AM"},"agent_actions":{"view_html":"https://pith.science/pith/DUCNE4AMJH2QPECADCQQBUC262","download_json":"https://pith.science/pith/DUCNE4AMJH2QPECADCQQBUC262.json","view_paper":"https://pith.science/paper/DUCNE4AM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1709.10089&json=true","fetch_graph":"https://pith.science/api/pith-number/DUCNE4AMJH2QPECADCQQBUC262/graph.json","fetch_events":"https://pith.science/api/pith-number/DUCNE4AMJH2QPECADCQQBUC262/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DUCNE4AMJH2QPECADCQQBUC262/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DUCNE4AMJH2QPECADCQQBUC262/action/storage_attestation","attest_author":"https://pith.science/pith/DUCNE4AMJH2QPECADCQQBUC262/action/author_attestation","sign_citation":"https://pith.science/pith/DUCNE4AMJH2QPECADCQQBUC262/action/citation_signature","submit_replication":"https://pith.science/pith/DUCNE4AMJH2QPECADCQQBUC262/action/replication_record"}},"created_at":"2026-05-18T00:22:39.464401+00:00","updated_at":"2026-05-18T00:22:39.464401+00:00"}