{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:YMPYQ7KXURBOI7PN2MP3CBHEWA","short_pith_number":"pith:YMPYQ7KX","canonical_record":{"source":{"id":"1903.08428","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2019-03-20T10:40:54Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"b905ae5fd8355bc8de74ff03b7a5904c52f408102195214610a970e6b1206eae","abstract_canon_sha256":"d2dd024e2d373a94c1c57ece2c089edd57bdb1f8fdad488206678e0dd4830fe4"},"schema_version":"1.0"},"canonical_sha256":"c31f887d57a442e47dedd31fb104e4b021f7dc02f77edb4f57691191f127500f","source":{"kind":"arxiv","id":"1903.08428","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1903.08428","created_at":"2026-05-17T23:50:44Z"},{"alias_kind":"arxiv_version","alias_value":"1903.08428v2","created_at":"2026-05-17T23:50:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1903.08428","created_at":"2026-05-17T23:50:44Z"},{"alias_kind":"pith_short_12","alias_value":"YMPYQ7KXURBO","created_at":"2026-05-18T12:33:33Z"},{"alias_kind":"pith_short_16","alias_value":"YMPYQ7KXURBOI7PN","created_at":"2026-05-18T12:33:33Z"},{"alias_kind":"pith_short_8","alias_value":"YMPYQ7KX","created_at":"2026-05-18T12:33:33Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:YMPYQ7KXURBOI7PN2MP3CBHEWA","target":"record","payload":{"canonical_record":{"source":{"id":"1903.08428","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2019-03-20T10:40:54Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"b905ae5fd8355bc8de74ff03b7a5904c52f408102195214610a970e6b1206eae","abstract_canon_sha256":"d2dd024e2d373a94c1c57ece2c089edd57bdb1f8fdad488206678e0dd4830fe4"},"schema_version":"1.0"},"canonical_sha256":"c31f887d57a442e47dedd31fb104e4b021f7dc02f77edb4f57691191f127500f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:50:44.326766Z","signature_b64":"Fkzt29L/3ixP1i2JexsOa3dT8M/sHhT0LrTIsgKc+fRYToLL4wLaBk3/ducoSJehqLglj1ZONsyDCArmko89Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c31f887d57a442e47dedd31fb104e4b021f7dc02f77edb4f57691191f127500f","last_reissued_at":"2026-05-17T23:50:44.326119Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:50:44.326119Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1903.08428","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:50:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"u7QMPrSM9AjN5x88hULtFQ1fbtKkL7FcZ8et6oJWhrQftN6rpJylC+69t3tEIDwX1LLnzyup0J3646Zr1A+1AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T23:30:46.843457Z"},"content_sha256":"76513fe2c9432febefa393e9d65e41fb7b4509b275afeb06142cb13dde29c497","schema_version":"1.0","event_id":"sha256:76513fe2c9432febefa393e9d65e41fb7b4509b275afeb06142cb13dde29c497"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:YMPYQ7KXURBOI7PN2MP3CBHEWA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Counterexample-Guided Strategy Improvement for POMDPs Using Recurrent Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Alexandru C. Serban, Bernd Becker, Nils Jansen, Ralf Wimmer, Steven Carr, Ufuk Topcu","submitted_at":"2019-03-20T10:40:54Z","abstract_excerpt":"We study strategy synthesis for partially observable Markov decision processes (POMDPs). The particular problem is to determine strategies that provably adhere to (probabilistic) temporal logic constraints. This problem is computationally intractable and theoretically hard. We propose a novel method that combines techniques from machine learning and formal verification. First, we train a recurrent neural network (RNN) to encode POMDP strategies. The RNN accounts for memory-based decisions without the need to expand the full belief space of a POMDP. Secondly, we restrict the RNN-based strategy "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1903.08428","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":""},"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":"2026-05-17T23:50:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mTRxAESRC20KZbAULQz2lXrmcbUnjlbLa0tKUyCftNazI3OJVz/9QVgH0FVSAfr37151B+6WQeX8nM3SvpmgAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T23:30:46.844171Z"},"content_sha256":"ec856e07588e7445b07a0834610efc5c384d3d92f515f052faaf28ab70530ac2","schema_version":"1.0","event_id":"sha256:ec856e07588e7445b07a0834610efc5c384d3d92f515f052faaf28ab70530ac2"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YMPYQ7KXURBOI7PN2MP3CBHEWA/bundle.json","state_url":"https://pith.science/pith/YMPYQ7KXURBOI7PN2MP3CBHEWA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YMPYQ7KXURBOI7PN2MP3CBHEWA/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":"2026-05-25T23:30:46Z","links":{"resolver":"https://pith.science/pith/YMPYQ7KXURBOI7PN2MP3CBHEWA","bundle":"https://pith.science/pith/YMPYQ7KXURBOI7PN2MP3CBHEWA/bundle.json","state":"https://pith.science/pith/YMPYQ7KXURBOI7PN2MP3CBHEWA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YMPYQ7KXURBOI7PN2MP3CBHEWA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:YMPYQ7KXURBOI7PN2MP3CBHEWA","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":"d2dd024e2d373a94c1c57ece2c089edd57bdb1f8fdad488206678e0dd4830fe4","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2019-03-20T10:40:54Z","title_canon_sha256":"b905ae5fd8355bc8de74ff03b7a5904c52f408102195214610a970e6b1206eae"},"schema_version":"1.0","source":{"id":"1903.08428","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1903.08428","created_at":"2026-05-17T23:50:44Z"},{"alias_kind":"arxiv_version","alias_value":"1903.08428v2","created_at":"2026-05-17T23:50:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1903.08428","created_at":"2026-05-17T23:50:44Z"},{"alias_kind":"pith_short_12","alias_value":"YMPYQ7KXURBO","created_at":"2026-05-18T12:33:33Z"},{"alias_kind":"pith_short_16","alias_value":"YMPYQ7KXURBOI7PN","created_at":"2026-05-18T12:33:33Z"},{"alias_kind":"pith_short_8","alias_value":"YMPYQ7KX","created_at":"2026-05-18T12:33:33Z"}],"graph_snapshots":[{"event_id":"sha256:ec856e07588e7445b07a0834610efc5c384d3d92f515f052faaf28ab70530ac2","target":"graph","created_at":"2026-05-17T23:50:44Z","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":"We study strategy synthesis for partially observable Markov decision processes (POMDPs). The particular problem is to determine strategies that provably adhere to (probabilistic) temporal logic constraints. This problem is computationally intractable and theoretically hard. We propose a novel method that combines techniques from machine learning and formal verification. First, we train a recurrent neural network (RNN) to encode POMDP strategies. The RNN accounts for memory-based decisions without the need to expand the full belief space of a POMDP. Secondly, we restrict the RNN-based strategy ","authors_text":"Alexandru C. Serban, Bernd Becker, Nils Jansen, Ralf Wimmer, Steven Carr, Ufuk Topcu","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2019-03-20T10:40:54Z","title":"Counterexample-Guided Strategy Improvement for POMDPs Using Recurrent Neural Networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1903.08428","kind":"arxiv","version":2},"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":"sha256:76513fe2c9432febefa393e9d65e41fb7b4509b275afeb06142cb13dde29c497","target":"record","created_at":"2026-05-17T23:50:44Z","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":"d2dd024e2d373a94c1c57ece2c089edd57bdb1f8fdad488206678e0dd4830fe4","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2019-03-20T10:40:54Z","title_canon_sha256":"b905ae5fd8355bc8de74ff03b7a5904c52f408102195214610a970e6b1206eae"},"schema_version":"1.0","source":{"id":"1903.08428","kind":"arxiv","version":2}},"canonical_sha256":"c31f887d57a442e47dedd31fb104e4b021f7dc02f77edb4f57691191f127500f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c31f887d57a442e47dedd31fb104e4b021f7dc02f77edb4f57691191f127500f","first_computed_at":"2026-05-17T23:50:44.326119Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:50:44.326119Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Fkzt29L/3ixP1i2JexsOa3dT8M/sHhT0LrTIsgKc+fRYToLL4wLaBk3/ducoSJehqLglj1ZONsyDCArmko89Bw==","signature_status":"signed_v1","signed_at":"2026-05-17T23:50:44.326766Z","signed_message":"canonical_sha256_bytes"},"source_id":"1903.08428","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:76513fe2c9432febefa393e9d65e41fb7b4509b275afeb06142cb13dde29c497","sha256:ec856e07588e7445b07a0834610efc5c384d3d92f515f052faaf28ab70530ac2"],"state_sha256":"16ab6b3aacdd674df8a76eb4a99e31bfee0c78f4dba8a2f10e535af696f778d1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3le1BG0CwM3zBghW2Rf2wEaBWZHpaa82e/aCXK4SMrT/+90HIRfyeiYwssBqXgW/uWf8MfixYQw6W1235HloCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-25T23:30:46.848243Z","bundle_sha256":"c96e8b75905a27f2fd5225f9a6817c95cafea7e05e9e9ea7e4b0f81f4bc775da"}}