{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:BYBXAEUBUDULDTYFFQ4JHQPMCB","short_pith_number":"pith:BYBXAEUB","schema_version":"1.0","canonical_sha256":"0e03701281a0e8b1cf052c3893c1ec105879cb32f8a71f48e61911168e802534","source":{"kind":"arxiv","id":"1901.06576","version":1},"attestation_state":"computed","paper":{"title":"Towards Physically Safe Reinforcement Learning under Supervision","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Devin Balkcom, Haoxiang Li, Yinan Zhang","submitted_at":"2019-01-19T19:16:42Z","abstract_excerpt":"This paper addresses the question of how a previously available control policy $\\pi_s$ can be used as a supervisor to more quickly and safely train a new learned control policy $\\pi_L$ for a robot. A weighted average of the supervisor and learned policies is used during trials, with a heavier weight initially on the supervisor, in order to allow safe and useful physical trials while the learned policy is still ineffective. During the process, the weight is adjusted to favor the learned policy. As weights are adjusted, the learned network must compensate so as to give safe and reasonable output"},"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":"1901.06576","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2019-01-19T19:16:42Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"d9e2b2fe213d75c8db6f842a43b69b039c5bc66800b91a93fa700b69e136f894","abstract_canon_sha256":"d9fa137e2ce9f448a0c6b93cc3a8a1de2c1183827ea625671c1bfa9af28817c1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:55:49.781177Z","signature_b64":"oeR5xIrJS4T8kdjGOQXk2HBB43m2jVLEy9JHWD8y5l14tcjpdamU6LQt8hKSJiYdGh4y+HAqIZotupXIuMcfCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0e03701281a0e8b1cf052c3893c1ec105879cb32f8a71f48e61911168e802534","last_reissued_at":"2026-05-17T23:55:49.780591Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:55:49.780591Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards Physically Safe Reinforcement Learning under Supervision","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Devin Balkcom, Haoxiang Li, Yinan Zhang","submitted_at":"2019-01-19T19:16:42Z","abstract_excerpt":"This paper addresses the question of how a previously available control policy $\\pi_s$ can be used as a supervisor to more quickly and safely train a new learned control policy $\\pi_L$ for a robot. A weighted average of the supervisor and learned policies is used during trials, with a heavier weight initially on the supervisor, in order to allow safe and useful physical trials while the learned policy is still ineffective. During the process, the weight is adjusted to favor the learned policy. As weights are adjusted, the learned network must compensate so as to give safe and reasonable output"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1901.06576","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":""},"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":"1901.06576","created_at":"2026-05-17T23:55:49.780693+00:00"},{"alias_kind":"arxiv_version","alias_value":"1901.06576v1","created_at":"2026-05-17T23:55:49.780693+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1901.06576","created_at":"2026-05-17T23:55:49.780693+00:00"},{"alias_kind":"pith_short_12","alias_value":"BYBXAEUBUDUL","created_at":"2026-05-18T12:33:12.712433+00:00"},{"alias_kind":"pith_short_16","alias_value":"BYBXAEUBUDULDTYF","created_at":"2026-05-18T12:33:12.712433+00:00"},{"alias_kind":"pith_short_8","alias_value":"BYBXAEUB","created_at":"2026-05-18T12:33:12.712433+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/BYBXAEUBUDULDTYFFQ4JHQPMCB","json":"https://pith.science/pith/BYBXAEUBUDULDTYFFQ4JHQPMCB.json","graph_json":"https://pith.science/api/pith-number/BYBXAEUBUDULDTYFFQ4JHQPMCB/graph.json","events_json":"https://pith.science/api/pith-number/BYBXAEUBUDULDTYFFQ4JHQPMCB/events.json","paper":"https://pith.science/paper/BYBXAEUB"},"agent_actions":{"view_html":"https://pith.science/pith/BYBXAEUBUDULDTYFFQ4JHQPMCB","download_json":"https://pith.science/pith/BYBXAEUBUDULDTYFFQ4JHQPMCB.json","view_paper":"https://pith.science/paper/BYBXAEUB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1901.06576&json=true","fetch_graph":"https://pith.science/api/pith-number/BYBXAEUBUDULDTYFFQ4JHQPMCB/graph.json","fetch_events":"https://pith.science/api/pith-number/BYBXAEUBUDULDTYFFQ4JHQPMCB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BYBXAEUBUDULDTYFFQ4JHQPMCB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BYBXAEUBUDULDTYFFQ4JHQPMCB/action/storage_attestation","attest_author":"https://pith.science/pith/BYBXAEUBUDULDTYFFQ4JHQPMCB/action/author_attestation","sign_citation":"https://pith.science/pith/BYBXAEUBUDULDTYFFQ4JHQPMCB/action/citation_signature","submit_replication":"https://pith.science/pith/BYBXAEUBUDULDTYFFQ4JHQPMCB/action/replication_record"}},"created_at":"2026-05-17T23:55:49.780693+00:00","updated_at":"2026-05-17T23:55:49.780693+00:00"}