{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:YELCLGCX4LIM6SQHJLQDYP6IY4","short_pith_number":"pith:YELCLGCX","schema_version":"1.0","canonical_sha256":"c116259857e2d0cf4a074ae03c3fc8c73d700b4b369e86bdb143f90d4d06899d","source":{"kind":"arxiv","id":"2106.13703","version":6},"attestation_state":"computed","paper":{"title":"Task-Driven Detection of Distribution Shifts with Statistical Guarantees for Robot Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.AP"],"primary_cat":"cs.RO","authors_text":"Alec Farid, Anirudha Majumdar, Divyanshu Pachisia, Sushant Veer","submitted_at":"2021-06-25T15:41:25Z","abstract_excerpt":"Our goal is to perform out-of-distribution (OOD) detection, i.e., to detect when a robot is operating in environments drawn from a different distribution than the ones used to train the robot. We leverage Probably Approximately Correct (PAC)-Bayes theory to train a policy with a guaranteed bound on performance on the training distribution. Our idea for OOD detection relies on the following intuition: violation of the performance bound on test environments provides evidence that the robot is operating OOD. We formalize this via statistical techniques based on p-values and concentration inequali"},"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":"2106.13703","kind":"arxiv","version":6},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2021-06-25T15:41:25Z","cross_cats_sorted":["cs.LG","stat.AP"],"title_canon_sha256":"3ba4b7c67c2d10df4e94a536dd13b7987d66bbfbf8dfc4ea45652c04d6757c4e","abstract_canon_sha256":"6e28a34e172a9a193c7ff2d0813f42d1c6c0f4d2a936a7b2327fe66294edb7bd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:09:41.561151Z","signature_b64":"Xz8AUyFNjq2x/+Ul13ICm1wWGIKTG5gugzWB/5FvHbJvr0aUA6hjC67SLPV6vbSlHVXSA1QlUymFKpFG3B2qCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c116259857e2d0cf4a074ae03c3fc8c73d700b4b369e86bdb143f90d4d06899d","last_reissued_at":"2026-07-05T07:09:41.560670Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:09:41.560670Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Task-Driven Detection of Distribution Shifts with Statistical Guarantees for Robot Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.AP"],"primary_cat":"cs.RO","authors_text":"Alec Farid, Anirudha Majumdar, Divyanshu Pachisia, Sushant Veer","submitted_at":"2021-06-25T15:41:25Z","abstract_excerpt":"Our goal is to perform out-of-distribution (OOD) detection, i.e., to detect when a robot is operating in environments drawn from a different distribution than the ones used to train the robot. We leverage Probably Approximately Correct (PAC)-Bayes theory to train a policy with a guaranteed bound on performance on the training distribution. Our idea for OOD detection relies on the following intuition: violation of the performance bound on test environments provides evidence that the robot is operating OOD. We formalize this via statistical techniques based on p-values and concentration inequali"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.13703","kind":"arxiv","version":6},"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/2106.13703/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":"2106.13703","created_at":"2026-07-05T07:09:41.560727+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.13703v6","created_at":"2026-07-05T07:09:41.560727+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.13703","created_at":"2026-07-05T07:09:41.560727+00:00"},{"alias_kind":"pith_short_12","alias_value":"YELCLGCX4LIM","created_at":"2026-07-05T07:09:41.560727+00:00"},{"alias_kind":"pith_short_16","alias_value":"YELCLGCX4LIM6SQH","created_at":"2026-07-05T07:09:41.560727+00:00"},{"alias_kind":"pith_short_8","alias_value":"YELCLGCX","created_at":"2026-07-05T07:09:41.560727+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.08508","citing_title":"ActProbe: Action-Space Probe for Early Failure Detection of Generative Robot Policies","ref_index":17,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YELCLGCX4LIM6SQHJLQDYP6IY4","json":"https://pith.science/pith/YELCLGCX4LIM6SQHJLQDYP6IY4.json","graph_json":"https://pith.science/api/pith-number/YELCLGCX4LIM6SQHJLQDYP6IY4/graph.json","events_json":"https://pith.science/api/pith-number/YELCLGCX4LIM6SQHJLQDYP6IY4/events.json","paper":"https://pith.science/paper/YELCLGCX"},"agent_actions":{"view_html":"https://pith.science/pith/YELCLGCX4LIM6SQHJLQDYP6IY4","download_json":"https://pith.science/pith/YELCLGCX4LIM6SQHJLQDYP6IY4.json","view_paper":"https://pith.science/paper/YELCLGCX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.13703&json=true","fetch_graph":"https://pith.science/api/pith-number/YELCLGCX4LIM6SQHJLQDYP6IY4/graph.json","fetch_events":"https://pith.science/api/pith-number/YELCLGCX4LIM6SQHJLQDYP6IY4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YELCLGCX4LIM6SQHJLQDYP6IY4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YELCLGCX4LIM6SQHJLQDYP6IY4/action/storage_attestation","attest_author":"https://pith.science/pith/YELCLGCX4LIM6SQHJLQDYP6IY4/action/author_attestation","sign_citation":"https://pith.science/pith/YELCLGCX4LIM6SQHJLQDYP6IY4/action/citation_signature","submit_replication":"https://pith.science/pith/YELCLGCX4LIM6SQHJLQDYP6IY4/action/replication_record"}},"created_at":"2026-07-05T07:09:41.560727+00:00","updated_at":"2026-07-05T07:09:41.560727+00:00"}