{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2017:77SJDTMALQZ7KU3ZOJYGQMSCQ5","short_pith_number":"pith:77SJDTMA","canonical_record":{"source":{"id":"1705.08551","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2017-05-23T22:20:08Z","cross_cats_sorted":["cs.AI","cs.LG","cs.SY"],"title_canon_sha256":"3280edaa7cd419ba1b26a1d690b246eaa81137a7d9075e1053fe5ccf7532368f","abstract_canon_sha256":"bf780f31d568051329e71cc62f3af2947fe5f724e1604012993c8165b0f7c607"},"schema_version":"1.0"},"canonical_sha256":"ffe491cd805c33f553797270683242877536cb99d092c3b58476bbc0c5db0ab1","source":{"kind":"arxiv","id":"1705.08551","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1705.08551","created_at":"2026-05-18T00:30:44Z"},{"alias_kind":"arxiv_version","alias_value":"1705.08551v3","created_at":"2026-05-18T00:30:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1705.08551","created_at":"2026-05-18T00:30:44Z"},{"alias_kind":"pith_short_12","alias_value":"77SJDTMALQZ7","created_at":"2026-05-18T12:31:03Z"},{"alias_kind":"pith_short_16","alias_value":"77SJDTMALQZ7KU3Z","created_at":"2026-05-18T12:31:03Z"},{"alias_kind":"pith_short_8","alias_value":"77SJDTMA","created_at":"2026-05-18T12:31:03Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2017:77SJDTMALQZ7KU3ZOJYGQMSCQ5","target":"record","payload":{"canonical_record":{"source":{"id":"1705.08551","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2017-05-23T22:20:08Z","cross_cats_sorted":["cs.AI","cs.LG","cs.SY"],"title_canon_sha256":"3280edaa7cd419ba1b26a1d690b246eaa81137a7d9075e1053fe5ccf7532368f","abstract_canon_sha256":"bf780f31d568051329e71cc62f3af2947fe5f724e1604012993c8165b0f7c607"},"schema_version":"1.0"},"canonical_sha256":"ffe491cd805c33f553797270683242877536cb99d092c3b58476bbc0c5db0ab1","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:30:44.677474Z","signature_b64":"aDIhTqCX9sMx7k3zeTafntRm3obfZL5DTdHSERQ19frG9kobhtyryYYZCuqietJStCN++Cx9cfaUP9g/+oL0Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ffe491cd805c33f553797270683242877536cb99d092c3b58476bbc0c5db0ab1","last_reissued_at":"2026-05-18T00:30:44.676902Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:30:44.676902Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1705.08551","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:30:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ERPXLYQd5PXWI0aRCMXvd+QMT/1bRPyZi0bZmulTyqWK8SSdfauOoFIvdyxKRz8/SEm9TMmFSYCQZnDKBoimBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T08:07:14.182117Z"},"content_sha256":"841f6f2e4f8393478a869c7132b14c923a0887781ab8d0137b391ffda792cde8","schema_version":"1.0","event_id":"sha256:841f6f2e4f8393478a869c7132b14c923a0887781ab8d0137b391ffda792cde8"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2017:77SJDTMALQZ7KU3ZOJYGQMSCQ5","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Safe Model-based Reinforcement Learning with Stability Guarantees","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.SY"],"primary_cat":"stat.ML","authors_text":"Andreas Krause, Angela P. Schoellig, Felix Berkenkamp, Matteo Turchetta","submitted_at":"2017-05-23T22:20:08Z","abstract_excerpt":"Reinforcement learning is a powerful paradigm for learning optimal policies from experimental data. However, to find optimal policies, most reinforcement learning algorithms explore all possible actions, which may be harmful for real-world systems. As a consequence, learning algorithms are rarely applied on safety-critical systems in the real world. In this paper, we present a learning algorithm that explicitly considers safety, defined in terms of stability guarantees. Specifically, we extend control-theoretic results on Lyapunov stability verification and show how to use statistical models o"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1705.08551","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":""},"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:30:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qCPTaCcU8UaFLZQVY+wLpHZ45e+QEXdt2P5lYab5oUrMCeYqHOKwZf3ZaWH+75T31WEcuMkhmDLasecpr+s/Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T08:07:14.182708Z"},"content_sha256":"e7f2f6a486456c5199a887bb67fd96ec8a7867be00d2cf92f31029f00a14731c","schema_version":"1.0","event_id":"sha256:e7f2f6a486456c5199a887bb67fd96ec8a7867be00d2cf92f31029f00a14731c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/77SJDTMALQZ7KU3ZOJYGQMSCQ5/bundle.json","state_url":"https://pith.science/pith/77SJDTMALQZ7KU3ZOJYGQMSCQ5/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/77SJDTMALQZ7KU3ZOJYGQMSCQ5/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-25T08:07:14Z","links":{"resolver":"https://pith.science/pith/77SJDTMALQZ7KU3ZOJYGQMSCQ5","bundle":"https://pith.science/pith/77SJDTMALQZ7KU3ZOJYGQMSCQ5/bundle.json","state":"https://pith.science/pith/77SJDTMALQZ7KU3ZOJYGQMSCQ5/state.json","well_known_bundle":"https://pith.science/.well-known/pith/77SJDTMALQZ7KU3ZOJYGQMSCQ5/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2017:77SJDTMALQZ7KU3ZOJYGQMSCQ5","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"bf780f31d568051329e71cc62f3af2947fe5f724e1604012993c8165b0f7c607","cross_cats_sorted":["cs.AI","cs.LG","cs.SY"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2017-05-23T22:20:08Z","title_canon_sha256":"3280edaa7cd419ba1b26a1d690b246eaa81137a7d9075e1053fe5ccf7532368f"},"schema_version":"1.0","source":{"id":"1705.08551","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1705.08551","created_at":"2026-05-18T00:30:44Z"},{"alias_kind":"arxiv_version","alias_value":"1705.08551v3","created_at":"2026-05-18T00:30:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1705.08551","created_at":"2026-05-18T00:30:44Z"},{"alias_kind":"pith_short_12","alias_value":"77SJDTMALQZ7","created_at":"2026-05-18T12:31:03Z"},{"alias_kind":"pith_short_16","alias_value":"77SJDTMALQZ7KU3Z","created_at":"2026-05-18T12:31:03Z"},{"alias_kind":"pith_short_8","alias_value":"77SJDTMA","created_at":"2026-05-18T12:31:03Z"}],"graph_snapshots":[{"event_id":"sha256:e7f2f6a486456c5199a887bb67fd96ec8a7867be00d2cf92f31029f00a14731c","target":"graph","created_at":"2026-05-18T00:30:44Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"Reinforcement learning is a powerful paradigm for learning optimal policies from experimental data. However, to find optimal policies, most reinforcement learning algorithms explore all possible actions, which may be harmful for real-world systems. As a consequence, learning algorithms are rarely applied on safety-critical systems in the real world. In this paper, we present a learning algorithm that explicitly considers safety, defined in terms of stability guarantees. Specifically, we extend control-theoretic results on Lyapunov stability verification and show how to use statistical models o","authors_text":"Andreas Krause, Angela P. Schoellig, Felix Berkenkamp, Matteo Turchetta","cross_cats":["cs.AI","cs.LG","cs.SY"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2017-05-23T22:20:08Z","title":"Safe Model-based Reinforcement Learning with Stability Guarantees"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1705.08551","kind":"arxiv","version":3},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:841f6f2e4f8393478a869c7132b14c923a0887781ab8d0137b391ffda792cde8","target":"record","created_at":"2026-05-18T00:30:44Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"bf780f31d568051329e71cc62f3af2947fe5f724e1604012993c8165b0f7c607","cross_cats_sorted":["cs.AI","cs.LG","cs.SY"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2017-05-23T22:20:08Z","title_canon_sha256":"3280edaa7cd419ba1b26a1d690b246eaa81137a7d9075e1053fe5ccf7532368f"},"schema_version":"1.0","source":{"id":"1705.08551","kind":"arxiv","version":3}},"canonical_sha256":"ffe491cd805c33f553797270683242877536cb99d092c3b58476bbc0c5db0ab1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ffe491cd805c33f553797270683242877536cb99d092c3b58476bbc0c5db0ab1","first_computed_at":"2026-05-18T00:30:44.676902Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:30:44.676902Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"aDIhTqCX9sMx7k3zeTafntRm3obfZL5DTdHSERQ19frG9kobhtyryYYZCuqietJStCN++Cx9cfaUP9g/+oL0Aw==","signature_status":"signed_v1","signed_at":"2026-05-18T00:30:44.677474Z","signed_message":"canonical_sha256_bytes"},"source_id":"1705.08551","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:841f6f2e4f8393478a869c7132b14c923a0887781ab8d0137b391ffda792cde8","sha256:e7f2f6a486456c5199a887bb67fd96ec8a7867be00d2cf92f31029f00a14731c"],"state_sha256":"0a819545bd20e56ebd9a616cc5c37a1b797b6bb09ec1052ff57eecaed5fb9779"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dHITdh8DtfTP8HrChtX/Nb2vhECWTaZRrObHp/uvoR/AMmWcYh9K3ZQ+SEXvs/IuFB22ub+wldcq0ZDQjWQGCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-25T08:07:14.185855Z","bundle_sha256":"78d1c222f3534f58aae49b6c35258e4dd75a4f6e07ea2c7f270d5428a4f6ccc7"}}