{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:E5EL6UZ6FVG3Q6YPVEROHM6L64","short_pith_number":"pith:E5EL6UZ6","schema_version":"1.0","canonical_sha256":"2748bf533e2d4db87b0fa922e3b3cbf724ce50929f9142be963ecf5e6fc65e35","source":{"kind":"arxiv","id":"2404.15199","version":3},"attestation_state":"computed","paper":{"title":"Reinforcement Learning with Adaptive Regularization for Safe Control of Critical Systems","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Haozhe Tian, Homayoun Hamedmoghadam, Pietro Ferraro, Robert Shorten","submitted_at":"2024-04-23T16:35:14Z","abstract_excerpt":"Reinforcement Learning (RL) is a powerful method for controlling dynamic systems, but its learning mechanism can lead to unpredictable actions that undermine the safety of critical systems. Here, we propose RL with Adaptive Regularization (RL-AR), an algorithm that enables safe RL exploration by combining the RL policy with a policy regularizer that hard-codes the safety constraints. RL-AR performs policy combination via a \"focus module,\" which determines the appropriate combination depending on the state--relying more on the safe policy regularizer for less-exploited states while allowing unb"},"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":"2404.15199","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-04-23T16:35:14Z","cross_cats_sorted":[],"title_canon_sha256":"6c05938b6aa01feca68956056019a8e2bdbf686322d8b6f806b00513ef675498","abstract_canon_sha256":"62837f9b5eff7b5e23566bc67e57eea0b5ce97c86945d38ee7ea04f3e6c77616"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:28:49.271645Z","signature_b64":"0/7h8E1/zwIWgFR34E83Grwt9Zg8okIVBp9iDC9Fl9Lg9/ivMnLWPJKQxKy8piGN2GWyTT2RSiiy1rrjWFQTAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2748bf533e2d4db87b0fa922e3b3cbf724ce50929f9142be963ecf5e6fc65e35","last_reissued_at":"2026-07-05T09:28:49.271155Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:28:49.271155Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reinforcement Learning with Adaptive Regularization for Safe Control of Critical Systems","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Haozhe Tian, Homayoun Hamedmoghadam, Pietro Ferraro, Robert Shorten","submitted_at":"2024-04-23T16:35:14Z","abstract_excerpt":"Reinforcement Learning (RL) is a powerful method for controlling dynamic systems, but its learning mechanism can lead to unpredictable actions that undermine the safety of critical systems. Here, we propose RL with Adaptive Regularization (RL-AR), an algorithm that enables safe RL exploration by combining the RL policy with a policy regularizer that hard-codes the safety constraints. RL-AR performs policy combination via a \"focus module,\" which determines the appropriate combination depending on the state--relying more on the safe policy regularizer for less-exploited states while allowing unb"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.15199","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/2404.15199/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":"2404.15199","created_at":"2026-07-05T09:28:49.271214+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.15199v3","created_at":"2026-07-05T09:28:49.271214+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.15199","created_at":"2026-07-05T09:28:49.271214+00:00"},{"alias_kind":"pith_short_12","alias_value":"E5EL6UZ6FVG3","created_at":"2026-07-05T09:28:49.271214+00:00"},{"alias_kind":"pith_short_16","alias_value":"E5EL6UZ6FVG3Q6YP","created_at":"2026-07-05T09:28:49.271214+00:00"},{"alias_kind":"pith_short_8","alias_value":"E5EL6UZ6","created_at":"2026-07-05T09:28:49.271214+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.31320","citing_title":"Safe Online Learning via Smooth Safety-Structured Policy Composition","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2510.11491","citing_title":"Constraint-Aware Reinforcement Learning via Adaptive Action Scaling","ref_index":22,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/E5EL6UZ6FVG3Q6YPVEROHM6L64","json":"https://pith.science/pith/E5EL6UZ6FVG3Q6YPVEROHM6L64.json","graph_json":"https://pith.science/api/pith-number/E5EL6UZ6FVG3Q6YPVEROHM6L64/graph.json","events_json":"https://pith.science/api/pith-number/E5EL6UZ6FVG3Q6YPVEROHM6L64/events.json","paper":"https://pith.science/paper/E5EL6UZ6"},"agent_actions":{"view_html":"https://pith.science/pith/E5EL6UZ6FVG3Q6YPVEROHM6L64","download_json":"https://pith.science/pith/E5EL6UZ6FVG3Q6YPVEROHM6L64.json","view_paper":"https://pith.science/paper/E5EL6UZ6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.15199&json=true","fetch_graph":"https://pith.science/api/pith-number/E5EL6UZ6FVG3Q6YPVEROHM6L64/graph.json","fetch_events":"https://pith.science/api/pith-number/E5EL6UZ6FVG3Q6YPVEROHM6L64/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/E5EL6UZ6FVG3Q6YPVEROHM6L64/action/timestamp_anchor","attest_storage":"https://pith.science/pith/E5EL6UZ6FVG3Q6YPVEROHM6L64/action/storage_attestation","attest_author":"https://pith.science/pith/E5EL6UZ6FVG3Q6YPVEROHM6L64/action/author_attestation","sign_citation":"https://pith.science/pith/E5EL6UZ6FVG3Q6YPVEROHM6L64/action/citation_signature","submit_replication":"https://pith.science/pith/E5EL6UZ6FVG3Q6YPVEROHM6L64/action/replication_record"}},"created_at":"2026-07-05T09:28:49.271214+00:00","updated_at":"2026-07-05T09:28:49.271214+00:00"}