{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:ZF44RWWHFSSRYKVJM66ZAR5O7G","short_pith_number":"pith:ZF44RWWH","schema_version":"1.0","canonical_sha256":"c979c8dac72ca51c2aa967bd9047aef9b64898d270968da227a696b0787266b6","source":{"kind":"arxiv","id":"2403.19648","version":2},"attestation_state":"computed","paper":{"title":"Human-compatible driving partners through data-regularized self-play reinforcement learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.MA"],"primary_cat":"cs.RO","authors_text":"Daphne Cornelisse, Eugene Vinitsky","submitted_at":"2024-03-28T17:56:56Z","abstract_excerpt":"A central challenge for autonomous vehicles is coordinating with humans. Therefore, incorporating realistic human agents is essential for scalable training and evaluation of autonomous driving systems in simulation. Simulation agents are typically developed by imitating large-scale, high-quality datasets of human driving. However, pure imitation learning agents empirically have high collision rates when executed in a multi-agent closed-loop setting. To build agents that are realistic and effective in closed-loop settings, we propose Human-Regularized PPO (HR-PPO), a multi-agent algorithm where"},"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":"2403.19648","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2024-03-28T17:56:56Z","cross_cats_sorted":["cs.AI","cs.LG","cs.MA"],"title_canon_sha256":"d62fd809e805502af17827a5d154ed447e5f94c7f9e6c35c386051fa65b9e3ee","abstract_canon_sha256":"117c6f2ccb5b292a3c41a63c7be625bdc68d0c6bd997dbd626304b98b401426d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:35:44.805881Z","signature_b64":"+sHszy9/Le3aIySm5NMoDpPveU416DbKiURtxIg758lhraLnNMLTCdC1l2QwGeAOgj0haRbnVTTA9LuwI2e6CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c979c8dac72ca51c2aa967bd9047aef9b64898d270968da227a696b0787266b6","last_reissued_at":"2026-07-05T08:35:44.805404Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:35:44.805404Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Human-compatible driving partners through data-regularized self-play reinforcement learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.MA"],"primary_cat":"cs.RO","authors_text":"Daphne Cornelisse, Eugene Vinitsky","submitted_at":"2024-03-28T17:56:56Z","abstract_excerpt":"A central challenge for autonomous vehicles is coordinating with humans. Therefore, incorporating realistic human agents is essential for scalable training and evaluation of autonomous driving systems in simulation. Simulation agents are typically developed by imitating large-scale, high-quality datasets of human driving. However, pure imitation learning agents empirically have high collision rates when executed in a multi-agent closed-loop setting. To build agents that are realistic and effective in closed-loop settings, we propose Human-Regularized PPO (HR-PPO), a multi-agent algorithm where"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.19648","kind":"arxiv","version":2},"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/2403.19648/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":"2403.19648","created_at":"2026-07-05T08:35:44.805462+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.19648v2","created_at":"2026-07-05T08:35:44.805462+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.19648","created_at":"2026-07-05T08:35:44.805462+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZF44RWWHFSSR","created_at":"2026-07-05T08:35:44.805462+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZF44RWWHFSSRYKVJ","created_at":"2026-07-05T08:35:44.805462+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZF44RWWH","created_at":"2026-07-05T08:35:44.805462+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.19370","citing_title":"Human-like autonomy emerges from self-play and a pinch of human data","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11688","citing_title":"Shaping Zero-Shot Coordination via State Blocking","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12857","citing_title":"Artificial Intelligence for Modeling and Simulation of Mixed Automated and Human Traffic","ref_index":109,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZF44RWWHFSSRYKVJM66ZAR5O7G","json":"https://pith.science/pith/ZF44RWWHFSSRYKVJM66ZAR5O7G.json","graph_json":"https://pith.science/api/pith-number/ZF44RWWHFSSRYKVJM66ZAR5O7G/graph.json","events_json":"https://pith.science/api/pith-number/ZF44RWWHFSSRYKVJM66ZAR5O7G/events.json","paper":"https://pith.science/paper/ZF44RWWH"},"agent_actions":{"view_html":"https://pith.science/pith/ZF44RWWHFSSRYKVJM66ZAR5O7G","download_json":"https://pith.science/pith/ZF44RWWHFSSRYKVJM66ZAR5O7G.json","view_paper":"https://pith.science/paper/ZF44RWWH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.19648&json=true","fetch_graph":"https://pith.science/api/pith-number/ZF44RWWHFSSRYKVJM66ZAR5O7G/graph.json","fetch_events":"https://pith.science/api/pith-number/ZF44RWWHFSSRYKVJM66ZAR5O7G/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZF44RWWHFSSRYKVJM66ZAR5O7G/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZF44RWWHFSSRYKVJM66ZAR5O7G/action/storage_attestation","attest_author":"https://pith.science/pith/ZF44RWWHFSSRYKVJM66ZAR5O7G/action/author_attestation","sign_citation":"https://pith.science/pith/ZF44RWWHFSSRYKVJM66ZAR5O7G/action/citation_signature","submit_replication":"https://pith.science/pith/ZF44RWWHFSSRYKVJM66ZAR5O7G/action/replication_record"}},"created_at":"2026-07-05T08:35:44.805462+00:00","updated_at":"2026-07-05T08:35:44.805462+00:00"}