{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:G7BWW4IZRVAN7SWPQM53BFQZ5W","short_pith_number":"pith:G7BWW4IZ","schema_version":"1.0","canonical_sha256":"37c36b71198d40dfcacf833bb09619edab0caa4071536abfe14eb2a0b6bfacf0","source":{"kind":"arxiv","id":"2010.10181","version":3},"attestation_state":"computed","paper":{"title":"Robust Imitation Learning from Noisy Demonstrations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"stat.ML","authors_text":"Masashi Sugiyama, Nontawat Charoenphakdee, Voot Tangkaratt","submitted_at":"2020-10-20T10:41:37Z","abstract_excerpt":"Robust learning from noisy demonstrations is a practical but highly challenging problem in imitation learning. In this paper, we first theoretically show that robust imitation learning can be achieved by optimizing a classification risk with a symmetric loss. Based on this theoretical finding, we then propose a new imitation learning method that optimizes the classification risk by effectively combining pseudo-labeling with co-training. Unlike existing methods, our method does not require additional labels or strict assumptions about noise distributions. Experimental results on continuous-cont"},"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":"2010.10181","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2020-10-20T10:41:37Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"d01fd2ec83fb871ce7fe048a0392d30cf6c6db1545f305599bfe8f5134f96243","abstract_canon_sha256":"0de730eead4f05e4e9253a23da8843f36082245a1fa557a8640e88e5c38610df"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:16:27.888536Z","signature_b64":"XhwGkbqzaF9L0R5cr8+43sG4tMpO73e7+SZR2rTfTbdVJDHvnjvwp5Qefu6KIEfyN0vUe41TNnmoQc2cqAoMCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"37c36b71198d40dfcacf833bb09619edab0caa4071536abfe14eb2a0b6bfacf0","last_reissued_at":"2026-07-05T02:16:27.888192Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:16:27.888192Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Robust Imitation Learning from Noisy Demonstrations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"stat.ML","authors_text":"Masashi Sugiyama, Nontawat Charoenphakdee, Voot Tangkaratt","submitted_at":"2020-10-20T10:41:37Z","abstract_excerpt":"Robust learning from noisy demonstrations is a practical but highly challenging problem in imitation learning. In this paper, we first theoretically show that robust imitation learning can be achieved by optimizing a classification risk with a symmetric loss. Based on this theoretical finding, we then propose a new imitation learning method that optimizes the classification risk by effectively combining pseudo-labeling with co-training. Unlike existing methods, our method does not require additional labels or strict assumptions about noise distributions. Experimental results on continuous-cont"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.10181","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/2010.10181/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":"2010.10181","created_at":"2026-07-05T02:16:27.888244+00:00"},{"alias_kind":"arxiv_version","alias_value":"2010.10181v3","created_at":"2026-07-05T02:16:27.888244+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.10181","created_at":"2026-07-05T02:16:27.888244+00:00"},{"alias_kind":"pith_short_12","alias_value":"G7BWW4IZRVAN","created_at":"2026-07-05T02:16:27.888244+00:00"},{"alias_kind":"pith_short_16","alias_value":"G7BWW4IZRVAN7SWP","created_at":"2026-07-05T02:16:27.888244+00:00"},{"alias_kind":"pith_short_8","alias_value":"G7BWW4IZ","created_at":"2026-07-05T02:16:27.888244+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.12365","citing_title":"Ambient Diffusion Policy: Imitation Learning from Suboptimal Data in Robotics","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13548","citing_title":"AttenA+: Rectifying Action Inequality in Robotic Foundation Models","ref_index":33,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/G7BWW4IZRVAN7SWPQM53BFQZ5W","json":"https://pith.science/pith/G7BWW4IZRVAN7SWPQM53BFQZ5W.json","graph_json":"https://pith.science/api/pith-number/G7BWW4IZRVAN7SWPQM53BFQZ5W/graph.json","events_json":"https://pith.science/api/pith-number/G7BWW4IZRVAN7SWPQM53BFQZ5W/events.json","paper":"https://pith.science/paper/G7BWW4IZ"},"agent_actions":{"view_html":"https://pith.science/pith/G7BWW4IZRVAN7SWPQM53BFQZ5W","download_json":"https://pith.science/pith/G7BWW4IZRVAN7SWPQM53BFQZ5W.json","view_paper":"https://pith.science/paper/G7BWW4IZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2010.10181&json=true","fetch_graph":"https://pith.science/api/pith-number/G7BWW4IZRVAN7SWPQM53BFQZ5W/graph.json","fetch_events":"https://pith.science/api/pith-number/G7BWW4IZRVAN7SWPQM53BFQZ5W/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/G7BWW4IZRVAN7SWPQM53BFQZ5W/action/timestamp_anchor","attest_storage":"https://pith.science/pith/G7BWW4IZRVAN7SWPQM53BFQZ5W/action/storage_attestation","attest_author":"https://pith.science/pith/G7BWW4IZRVAN7SWPQM53BFQZ5W/action/author_attestation","sign_citation":"https://pith.science/pith/G7BWW4IZRVAN7SWPQM53BFQZ5W/action/citation_signature","submit_replication":"https://pith.science/pith/G7BWW4IZRVAN7SWPQM53BFQZ5W/action/replication_record"}},"created_at":"2026-07-05T02:16:27.888244+00:00","updated_at":"2026-07-05T02:16:27.888244+00:00"}