{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:JJCVBABK5UFTQPOYDUMZUH5BCE","short_pith_number":"pith:JJCVBABK","schema_version":"1.0","canonical_sha256":"4a4550802aed0b383dd81d199a1fa1112865e5f3c6f6c037b7427027de9d9c2c","source":{"kind":"arxiv","id":"2110.14468","version":3},"attestation_state":"computed","paper":{"title":"DESTA: A Framework for Safe Reinforcement Learning with Markov Games of Intervention","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Aivar Sootla, Changmin Yu, David Mguni, Joel Jennings, Jun Wang, Usman Islam, Xiuling Zhang, Yaodong Yang, Yaqi Sun, Ziyan Wang","submitted_at":"2021-10-27T14:35:00Z","abstract_excerpt":"Reinforcement learning (RL) involves performing exploratory actions in an unknown system. This can place a learning agent in dangerous and potentially catastrophic system states. Current approaches for tackling safe learning in RL simultaneously trade-off safe exploration and task fulfillment. In this paper, we introduce a new generation of RL solvers that learn to minimise safety violations while maximising the task reward to the extent that can be tolerated by the safe policy. Our approach introduces a novel two-player framework for safe RL called Distributive Exploration Safety Training Alg"},"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":"2110.14468","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-10-27T14:35:00Z","cross_cats_sorted":[],"title_canon_sha256":"f67bf81ba4caaf1582dc22deac353c3e650e7d0651cfb64bf975f451a2e8d492","abstract_canon_sha256":"35c3a0b55f674c1aaabc05b3c8421aeb47ea9a394be45e0f82365ed60878401d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:47:01.208953Z","signature_b64":"KRvXn3KL4dJXnNGEzzBbfKMh3LvWBlfMjHSlb75vp5hvgxioy7k0S7kpS+rHtWaQPlA0nseaeT1kkPc6ynP+Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4a4550802aed0b383dd81d199a1fa1112865e5f3c6f6c037b7427027de9d9c2c","last_reissued_at":"2026-07-05T05:47:01.208548Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:47:01.208548Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DESTA: A Framework for Safe Reinforcement Learning with Markov Games of Intervention","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Aivar Sootla, Changmin Yu, David Mguni, Joel Jennings, Jun Wang, Usman Islam, Xiuling Zhang, Yaodong Yang, Yaqi Sun, Ziyan Wang","submitted_at":"2021-10-27T14:35:00Z","abstract_excerpt":"Reinforcement learning (RL) involves performing exploratory actions in an unknown system. This can place a learning agent in dangerous and potentially catastrophic system states. Current approaches for tackling safe learning in RL simultaneously trade-off safe exploration and task fulfillment. In this paper, we introduce a new generation of RL solvers that learn to minimise safety violations while maximising the task reward to the extent that can be tolerated by the safe policy. Our approach introduces a novel two-player framework for safe RL called Distributive Exploration Safety Training Alg"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.14468","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/2110.14468/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":"2110.14468","created_at":"2026-07-05T05:47:01.208606+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.14468v3","created_at":"2026-07-05T05:47:01.208606+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.14468","created_at":"2026-07-05T05:47:01.208606+00:00"},{"alias_kind":"pith_short_12","alias_value":"JJCVBABK5UFT","created_at":"2026-07-05T05:47:01.208606+00:00"},{"alias_kind":"pith_short_16","alias_value":"JJCVBABK5UFTQPOY","created_at":"2026-07-05T05:47:01.208606+00:00"},{"alias_kind":"pith_short_8","alias_value":"JJCVBABK","created_at":"2026-07-05T05:47:01.208606+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/JJCVBABK5UFTQPOYDUMZUH5BCE","json":"https://pith.science/pith/JJCVBABK5UFTQPOYDUMZUH5BCE.json","graph_json":"https://pith.science/api/pith-number/JJCVBABK5UFTQPOYDUMZUH5BCE/graph.json","events_json":"https://pith.science/api/pith-number/JJCVBABK5UFTQPOYDUMZUH5BCE/events.json","paper":"https://pith.science/paper/JJCVBABK"},"agent_actions":{"view_html":"https://pith.science/pith/JJCVBABK5UFTQPOYDUMZUH5BCE","download_json":"https://pith.science/pith/JJCVBABK5UFTQPOYDUMZUH5BCE.json","view_paper":"https://pith.science/paper/JJCVBABK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.14468&json=true","fetch_graph":"https://pith.science/api/pith-number/JJCVBABK5UFTQPOYDUMZUH5BCE/graph.json","fetch_events":"https://pith.science/api/pith-number/JJCVBABK5UFTQPOYDUMZUH5BCE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JJCVBABK5UFTQPOYDUMZUH5BCE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JJCVBABK5UFTQPOYDUMZUH5BCE/action/storage_attestation","attest_author":"https://pith.science/pith/JJCVBABK5UFTQPOYDUMZUH5BCE/action/author_attestation","sign_citation":"https://pith.science/pith/JJCVBABK5UFTQPOYDUMZUH5BCE/action/citation_signature","submit_replication":"https://pith.science/pith/JJCVBABK5UFTQPOYDUMZUH5BCE/action/replication_record"}},"created_at":"2026-07-05T05:47:01.208606+00:00","updated_at":"2026-07-05T05:47:01.208606+00:00"}