{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:UKXBZXKAV45EO2PDKTP327IXFS","short_pith_number":"pith:UKXBZXKA","schema_version":"1.0","canonical_sha256":"a2ae1cdd40af3a4769e354dfbd7d172ca53782b5c6a65fc67acc15d98f327dc9","source":{"kind":"arxiv","id":"1907.03976","version":3},"attestation_state":"computed","paper":{"title":"Better-than-Demonstrator Imitation Learning via Automatically-Ranked Demonstrations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Daniel S. Brown, Scott Niekum, Wonjoon Goo","submitted_at":"2019-07-09T04:11:53Z","abstract_excerpt":"The performance of imitation learning is typically upper-bounded by the performance of the demonstrator. While recent empirical results demonstrate that ranked demonstrations allow for better-than-demonstrator performance, preferences over demonstrations may be difficult to obtain, and little is known theoretically about when such methods can be expected to successfully extrapolate beyond the performance of the demonstrator. To address these issues, we first contribute a sufficient condition for better-than-demonstrator imitation learning and provide theoretical results showing why preferences"},"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":"1907.03976","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-07-09T04:11:53Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"25f4925b81287197e5e56dabd446da272de787ac4325f813a61ae2906d2d93a6","abstract_canon_sha256":"ead99042b08d264db327c867125fe26f6a158b6f6d5b1c708e68068f7319eb46"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:11:35.188748Z","signature_b64":"GFE5O/uuIai/GzRaOIhNxCijrlLaE4EC1Ugz1M/snDOU72tUA4HbSizjMWeZ48KeasP35D3j/JB6S+CWk1ZQCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a2ae1cdd40af3a4769e354dfbd7d172ca53782b5c6a65fc67acc15d98f327dc9","last_reissued_at":"2026-07-05T00:11:35.188206Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:11:35.188206Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Better-than-Demonstrator Imitation Learning via Automatically-Ranked Demonstrations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Daniel S. Brown, Scott Niekum, Wonjoon Goo","submitted_at":"2019-07-09T04:11:53Z","abstract_excerpt":"The performance of imitation learning is typically upper-bounded by the performance of the demonstrator. While recent empirical results demonstrate that ranked demonstrations allow for better-than-demonstrator performance, preferences over demonstrations may be difficult to obtain, and little is known theoretically about when such methods can be expected to successfully extrapolate beyond the performance of the demonstrator. To address these issues, we first contribute a sufficient condition for better-than-demonstrator imitation learning and provide theoretical results showing why preferences"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1907.03976","kind":"arxiv","version":3},"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/1907.03976/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":"1907.03976","created_at":"2026-07-05T00:11:35.188267+00:00"},{"alias_kind":"arxiv_version","alias_value":"1907.03976v3","created_at":"2026-07-05T00:11:35.188267+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1907.03976","created_at":"2026-07-05T00:11:35.188267+00:00"},{"alias_kind":"pith_short_12","alias_value":"UKXBZXKAV45E","created_at":"2026-07-05T00:11:35.188267+00:00"},{"alias_kind":"pith_short_16","alias_value":"UKXBZXKAV45EO2PD","created_at":"2026-07-05T00:11:35.188267+00:00"},{"alias_kind":"pith_short_8","alias_value":"UKXBZXKA","created_at":"2026-07-05T00:11:35.188267+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.22626","citing_title":"SCIZOR: A Self-Supervised Approach to Data Curation for Large-Scale Imitation Learning","ref_index":2,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UKXBZXKAV45EO2PDKTP327IXFS","json":"https://pith.science/pith/UKXBZXKAV45EO2PDKTP327IXFS.json","graph_json":"https://pith.science/api/pith-number/UKXBZXKAV45EO2PDKTP327IXFS/graph.json","events_json":"https://pith.science/api/pith-number/UKXBZXKAV45EO2PDKTP327IXFS/events.json","paper":"https://pith.science/paper/UKXBZXKA"},"agent_actions":{"view_html":"https://pith.science/pith/UKXBZXKAV45EO2PDKTP327IXFS","download_json":"https://pith.science/pith/UKXBZXKAV45EO2PDKTP327IXFS.json","view_paper":"https://pith.science/paper/UKXBZXKA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1907.03976&json=true","fetch_graph":"https://pith.science/api/pith-number/UKXBZXKAV45EO2PDKTP327IXFS/graph.json","fetch_events":"https://pith.science/api/pith-number/UKXBZXKAV45EO2PDKTP327IXFS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UKXBZXKAV45EO2PDKTP327IXFS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UKXBZXKAV45EO2PDKTP327IXFS/action/storage_attestation","attest_author":"https://pith.science/pith/UKXBZXKAV45EO2PDKTP327IXFS/action/author_attestation","sign_citation":"https://pith.science/pith/UKXBZXKAV45EO2PDKTP327IXFS/action/citation_signature","submit_replication":"https://pith.science/pith/UKXBZXKAV45EO2PDKTP327IXFS/action/replication_record"}},"created_at":"2026-07-05T00:11:35.188267+00:00","updated_at":"2026-07-05T00:11:35.188267+00:00"}