{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:4SLNCK7PXRZA7CGLHOILTZK3QV","short_pith_number":"pith:4SLNCK7P","schema_version":"1.0","canonical_sha256":"e496d12befbc720f88cb3b90b9e55b85452f7a218c1f36c5937bfd9bbc254fe1","source":{"kind":"arxiv","id":"2112.00950","version":1},"attestation_state":"computed","paper":{"title":"Quantile Filtered Imitation Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"David Brandfonbrener, Joan Bruna, Rajesh Ranganath, William F. Whitney","submitted_at":"2021-12-02T03:08:23Z","abstract_excerpt":"We introduce quantile filtered imitation learning (QFIL), a novel policy improvement operator designed for offline reinforcement learning. QFIL performs policy improvement by running imitation learning on a filtered version of the offline dataset. The filtering process removes $ s,a $ pairs whose estimated Q values fall below a given quantile of the pushforward distribution over values induced by sampling actions from the behavior policy. The definitions of both the pushforward Q distribution and resulting value function quantile are key contributions of our method. We prove that QFIL gives us"},"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":"2112.00950","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-12-02T03:08:23Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"ec0519715e3dac289ca44879b66f1f18ccc0c83587a13df55dfa59674eafa8d5","abstract_canon_sha256":"6d9b540f9f71b7d184f7ba96c3dd50c60c672555567700079dec0b540cfaec9e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:37:01.792689Z","signature_b64":"2j10bmwGtBg+Ks9g/zGmeI7cVz9bUIfS4iH1d6WoU8BUonZgQo8+0ZTAgO0iDeGr+7RZoZvkU7y8Ak8o4GU8Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e496d12befbc720f88cb3b90b9e55b85452f7a218c1f36c5937bfd9bbc254fe1","last_reissued_at":"2026-07-05T03:37:01.792230Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:37:01.792230Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Quantile Filtered Imitation Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"David Brandfonbrener, Joan Bruna, Rajesh Ranganath, William F. Whitney","submitted_at":"2021-12-02T03:08:23Z","abstract_excerpt":"We introduce quantile filtered imitation learning (QFIL), a novel policy improvement operator designed for offline reinforcement learning. QFIL performs policy improvement by running imitation learning on a filtered version of the offline dataset. The filtering process removes $ s,a $ pairs whose estimated Q values fall below a given quantile of the pushforward distribution over values induced by sampling actions from the behavior policy. The definitions of both the pushforward Q distribution and resulting value function quantile are key contributions of our method. We prove that QFIL gives us"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2112.00950","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/2112.00950/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":"2112.00950","created_at":"2026-07-05T03:37:01.792287+00:00"},{"alias_kind":"arxiv_version","alias_value":"2112.00950v1","created_at":"2026-07-05T03:37:01.792287+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2112.00950","created_at":"2026-07-05T03:37:01.792287+00:00"},{"alias_kind":"pith_short_12","alias_value":"4SLNCK7PXRZA","created_at":"2026-07-05T03:37:01.792287+00:00"},{"alias_kind":"pith_short_16","alias_value":"4SLNCK7PXRZA7CGL","created_at":"2026-07-05T03:37:01.792287+00:00"},{"alias_kind":"pith_short_8","alias_value":"4SLNCK7P","created_at":"2026-07-05T03:37:01.792287+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4SLNCK7PXRZA7CGLHOILTZK3QV","json":"https://pith.science/pith/4SLNCK7PXRZA7CGLHOILTZK3QV.json","graph_json":"https://pith.science/api/pith-number/4SLNCK7PXRZA7CGLHOILTZK3QV/graph.json","events_json":"https://pith.science/api/pith-number/4SLNCK7PXRZA7CGLHOILTZK3QV/events.json","paper":"https://pith.science/paper/4SLNCK7P"},"agent_actions":{"view_html":"https://pith.science/pith/4SLNCK7PXRZA7CGLHOILTZK3QV","download_json":"https://pith.science/pith/4SLNCK7PXRZA7CGLHOILTZK3QV.json","view_paper":"https://pith.science/paper/4SLNCK7P","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2112.00950&json=true","fetch_graph":"https://pith.science/api/pith-number/4SLNCK7PXRZA7CGLHOILTZK3QV/graph.json","fetch_events":"https://pith.science/api/pith-number/4SLNCK7PXRZA7CGLHOILTZK3QV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4SLNCK7PXRZA7CGLHOILTZK3QV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4SLNCK7PXRZA7CGLHOILTZK3QV/action/storage_attestation","attest_author":"https://pith.science/pith/4SLNCK7PXRZA7CGLHOILTZK3QV/action/author_attestation","sign_citation":"https://pith.science/pith/4SLNCK7PXRZA7CGLHOILTZK3QV/action/citation_signature","submit_replication":"https://pith.science/pith/4SLNCK7PXRZA7CGLHOILTZK3QV/action/replication_record"}},"created_at":"2026-07-05T03:37:01.792287+00:00","updated_at":"2026-07-05T03:37:01.792287+00:00"}