{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2016:YRTEFE4QUDW7VZBVGLEMT75UOV","short_pith_number":"pith:YRTEFE4Q","schema_version":"1.0","canonical_sha256":"c466429390a0edfae43532c8c9ffb475702ffe9858c6731e8d3660453ed40f96","source":{"kind":"arxiv","id":"1606.03137","version":4},"attestation_state":"computed","paper":{"title":"Cooperative Inverse Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Anca Dragan, Dylan Hadfield-Menell, Pieter Abbeel, Stuart Russell","submitted_at":"2016-06-09T22:39:54Z","abstract_excerpt":"For an autonomous system to be helpful to humans and to pose no unwarranted risks, it needs to align its values with those of the humans in its environment in such a way that its actions contribute to the maximization of value for the humans. We propose a formal definition of the value alignment problem as cooperative inverse reinforcement learning (CIRL). A CIRL problem is a cooperative, partial-information game with two agents, human and robot; both are rewarded according to the human's reward function, but the robot does not initially know what this is. In contrast to classical IRL, where t"},"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":"1606.03137","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2016-06-09T22:39:54Z","cross_cats_sorted":[],"title_canon_sha256":"a693c21211efef1dcf3c342e5b329007148c44be2880eb4cba458ade74f4d569","abstract_canon_sha256":"7a2e5d362748ebc84cc347ee01d9d5e1e3fdc5757746269e493df63451315610"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:46:08.402962Z","signature_b64":"1CkacsUtJZIC/+wJhHVWpgWwZ1qG0Ci+9ObznAFjj/SFp3uCbWu0t2wn0JPcDQXSH3Ter0xEUETwgNRPMzKqAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c466429390a0edfae43532c8c9ffb475702ffe9858c6731e8d3660453ed40f96","last_reissued_at":"2026-07-05T07:46:08.402590Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:46:08.402590Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Cooperative Inverse Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Anca Dragan, Dylan Hadfield-Menell, Pieter Abbeel, Stuart Russell","submitted_at":"2016-06-09T22:39:54Z","abstract_excerpt":"For an autonomous system to be helpful to humans and to pose no unwarranted risks, it needs to align its values with those of the humans in its environment in such a way that its actions contribute to the maximization of value for the humans. We propose a formal definition of the value alignment problem as cooperative inverse reinforcement learning (CIRL). A CIRL problem is a cooperative, partial-information game with two agents, human and robot; both are rewarded according to the human's reward function, but the robot does not initially know what this is. In contrast to classical IRL, where t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1606.03137","kind":"arxiv","version":4},"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/1606.03137/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":"1606.03137","created_at":"2026-07-05T07:46:08.402648+00:00"},{"alias_kind":"arxiv_version","alias_value":"1606.03137v4","created_at":"2026-07-05T07:46:08.402648+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1606.03137","created_at":"2026-07-05T07:46:08.402648+00:00"},{"alias_kind":"pith_short_12","alias_value":"YRTEFE4QUDW7","created_at":"2026-07-05T07:46:08.402648+00:00"},{"alias_kind":"pith_short_16","alias_value":"YRTEFE4QUDW7VZBV","created_at":"2026-07-05T07:46:08.402648+00:00"},{"alias_kind":"pith_short_8","alias_value":"YRTEFE4Q","created_at":"2026-07-05T07:46:08.402648+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.31572","citing_title":"FormIDEAble: Safe and Socially-aware Autonomous Systems","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"1907.00452","citing_title":"Detecting Spiky Corruption in Markov Decision Processes","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2604.23338","citing_title":"A Systematic Survey of Security Threats and Defenses in LLM-Based AI Agents: A Layered Attack Surface Framework","ref_index":132,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05851","citing_title":"Hypothesis generation and updating in large language models","ref_index":67,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YRTEFE4QUDW7VZBVGLEMT75UOV","json":"https://pith.science/pith/YRTEFE4QUDW7VZBVGLEMT75UOV.json","graph_json":"https://pith.science/api/pith-number/YRTEFE4QUDW7VZBVGLEMT75UOV/graph.json","events_json":"https://pith.science/api/pith-number/YRTEFE4QUDW7VZBVGLEMT75UOV/events.json","paper":"https://pith.science/paper/YRTEFE4Q"},"agent_actions":{"view_html":"https://pith.science/pith/YRTEFE4QUDW7VZBVGLEMT75UOV","download_json":"https://pith.science/pith/YRTEFE4QUDW7VZBVGLEMT75UOV.json","view_paper":"https://pith.science/paper/YRTEFE4Q","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1606.03137&json=true","fetch_graph":"https://pith.science/api/pith-number/YRTEFE4QUDW7VZBVGLEMT75UOV/graph.json","fetch_events":"https://pith.science/api/pith-number/YRTEFE4QUDW7VZBVGLEMT75UOV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YRTEFE4QUDW7VZBVGLEMT75UOV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YRTEFE4QUDW7VZBVGLEMT75UOV/action/storage_attestation","attest_author":"https://pith.science/pith/YRTEFE4QUDW7VZBVGLEMT75UOV/action/author_attestation","sign_citation":"https://pith.science/pith/YRTEFE4QUDW7VZBVGLEMT75UOV/action/citation_signature","submit_replication":"https://pith.science/pith/YRTEFE4QUDW7VZBVGLEMT75UOV/action/replication_record"}},"created_at":"2026-07-05T07:46:08.402648+00:00","updated_at":"2026-07-05T07:46:08.402648+00:00"}