{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:OK6LKBRCWJKG4LXTUIACYDUKIT","short_pith_number":"pith:OK6LKBRC","schema_version":"1.0","canonical_sha256":"72bcb50622b2546e2ef3a2002c0e8a44dc59c1a00f21bc05329e95f6dc34f95f","source":{"kind":"arxiv","id":"2009.05990","version":1},"attestation_state":"computed","paper":{"title":"Toward the Fundamental Limits of Imitation Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Jiantao Jiao, Kannan Ramachandran, Lin F. Yang, Nived Rajaraman","submitted_at":"2020-09-13T12:45:52Z","abstract_excerpt":"Imitation learning (IL) aims to mimic the behavior of an expert policy in a sequential decision-making problem given only demonstrations. In this paper, we focus on understanding the minimax statistical limits of IL in episodic Markov Decision Processes (MDPs). We first consider the setting where the learner is provided a dataset of $N$ expert trajectories ahead of time, and cannot interact with the MDP. Here, we show that the policy which mimics the expert whenever possible is in expectation $\\lesssim \\frac{|\\mathcal{S}| H^2 \\log (N)}{N}$ suboptimal compared to the value of the expert, even w"},"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":"2009.05990","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-09-13T12:45:52Z","cross_cats_sorted":["cs.AI","math.OC","stat.ML"],"title_canon_sha256":"3c772a7b4149d7deb761405d05ea3e0c93f8ad0857a1f685c8faf74c695301c0","abstract_canon_sha256":"6e5d2b7e00dae351f414f0a0057d31359fd790fad66c315afbc64c1118e2af59"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:35:03.321425Z","signature_b64":"pM0/q5zlBoFrui41WqnrA4wghn114KN4IkJ55CJxMmVWm9dyRFdbXSoxjAuYUeT5YCKaKHG44gOIzsjdqY6SBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"72bcb50622b2546e2ef3a2002c0e8a44dc59c1a00f21bc05329e95f6dc34f95f","last_reissued_at":"2026-07-05T01:35:03.321037Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:35:03.321037Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Toward the Fundamental Limits of Imitation Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Jiantao Jiao, Kannan Ramachandran, Lin F. Yang, Nived Rajaraman","submitted_at":"2020-09-13T12:45:52Z","abstract_excerpt":"Imitation learning (IL) aims to mimic the behavior of an expert policy in a sequential decision-making problem given only demonstrations. In this paper, we focus on understanding the minimax statistical limits of IL in episodic Markov Decision Processes (MDPs). We first consider the setting where the learner is provided a dataset of $N$ expert trajectories ahead of time, and cannot interact with the MDP. Here, we show that the policy which mimics the expert whenever possible is in expectation $\\lesssim \\frac{|\\mathcal{S}| H^2 \\log (N)}{N}$ suboptimal compared to the value of the expert, even w"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2009.05990","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/2009.05990/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":"2009.05990","created_at":"2026-07-05T01:35:03.321095+00:00"},{"alias_kind":"arxiv_version","alias_value":"2009.05990v1","created_at":"2026-07-05T01:35:03.321095+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2009.05990","created_at":"2026-07-05T01:35:03.321095+00:00"},{"alias_kind":"pith_short_12","alias_value":"OK6LKBRCWJKG","created_at":"2026-07-05T01:35:03.321095+00:00"},{"alias_kind":"pith_short_16","alias_value":"OK6LKBRCWJKG4LXT","created_at":"2026-07-05T01:35:03.321095+00:00"},{"alias_kind":"pith_short_8","alias_value":"OK6LKBRC","created_at":"2026-07-05T01:35:03.321095+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.21501","citing_title":"Few-Shot Neuro-Symbolic Imitation Learning for Long-Horizon Planning and Acting","ref_index":6,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OK6LKBRCWJKG4LXTUIACYDUKIT","json":"https://pith.science/pith/OK6LKBRCWJKG4LXTUIACYDUKIT.json","graph_json":"https://pith.science/api/pith-number/OK6LKBRCWJKG4LXTUIACYDUKIT/graph.json","events_json":"https://pith.science/api/pith-number/OK6LKBRCWJKG4LXTUIACYDUKIT/events.json","paper":"https://pith.science/paper/OK6LKBRC"},"agent_actions":{"view_html":"https://pith.science/pith/OK6LKBRCWJKG4LXTUIACYDUKIT","download_json":"https://pith.science/pith/OK6LKBRCWJKG4LXTUIACYDUKIT.json","view_paper":"https://pith.science/paper/OK6LKBRC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2009.05990&json=true","fetch_graph":"https://pith.science/api/pith-number/OK6LKBRCWJKG4LXTUIACYDUKIT/graph.json","fetch_events":"https://pith.science/api/pith-number/OK6LKBRCWJKG4LXTUIACYDUKIT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OK6LKBRCWJKG4LXTUIACYDUKIT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OK6LKBRCWJKG4LXTUIACYDUKIT/action/storage_attestation","attest_author":"https://pith.science/pith/OK6LKBRCWJKG4LXTUIACYDUKIT/action/author_attestation","sign_citation":"https://pith.science/pith/OK6LKBRCWJKG4LXTUIACYDUKIT/action/citation_signature","submit_replication":"https://pith.science/pith/OK6LKBRCWJKG4LXTUIACYDUKIT/action/replication_record"}},"created_at":"2026-07-05T01:35:03.321095+00:00","updated_at":"2026-07-05T01:35:03.321095+00:00"}