{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:3Z3GZJAPPJHNU3EFA3MN5AFEWV","short_pith_number":"pith:3Z3GZJAP","schema_version":"1.0","canonical_sha256":"de766ca40f7a4eda6c8506d8de80a4b5641b4245548df9cdf1d27f6a366d69c6","source":{"kind":"arxiv","id":"1909.11591","version":2},"attestation_state":"computed","paper":{"title":"Modular Deep Reinforcement Learning with Temporal Logic Specifications","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LO","cs.SY","eess.SY","stat.ML"],"primary_cat":"cs.LG","authors_text":"Alessandro Abate, Daniel Kroening, Lim Zun Yuan, Mohammadhosein Hasanbeig","submitted_at":"2019-09-23T18:10:00Z","abstract_excerpt":"We propose an actor-critic, model-free, and online Reinforcement Learning (RL) framework for continuous-state continuous-action Markov Decision Processes (MDPs) when the reward is highly sparse but encompasses a high-level temporal structure. We represent this temporal structure by a finite-state machine and construct an on-the-fly synchronised product with the MDP and the finite machine. The temporal structure acts as a guide for the RL agent within the product, where a modular Deep Deterministic Policy Gradient (DDPG) architecture is proposed to generate a low-level control policy. We evalua"},"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":"1909.11591","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-09-23T18:10:00Z","cross_cats_sorted":["cs.AI","cs.LO","cs.SY","eess.SY","stat.ML"],"title_canon_sha256":"1dd7195475b7ded53764122ad4f14270cb208a76d437c513c36c75c5a537a36c","abstract_canon_sha256":"5b46bcad2fabd122c2b892ed8baddebcf90342e037eea5f166da3777af2ae6ad"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:21:37.178491Z","signature_b64":"AoBox9LQd6kLvZHOc+WawqHls5ljE7VLNzeLYh7pZLiL+doltcR7FIEcvpRSAMdjgdzuUPj7P9vXNEuB4ZOSBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"de766ca40f7a4eda6c8506d8de80a4b5641b4245548df9cdf1d27f6a366d69c6","last_reissued_at":"2026-07-05T00:21:37.177933Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:21:37.177933Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Modular Deep Reinforcement Learning with Temporal Logic Specifications","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LO","cs.SY","eess.SY","stat.ML"],"primary_cat":"cs.LG","authors_text":"Alessandro Abate, Daniel Kroening, Lim Zun Yuan, Mohammadhosein Hasanbeig","submitted_at":"2019-09-23T18:10:00Z","abstract_excerpt":"We propose an actor-critic, model-free, and online Reinforcement Learning (RL) framework for continuous-state continuous-action Markov Decision Processes (MDPs) when the reward is highly sparse but encompasses a high-level temporal structure. We represent this temporal structure by a finite-state machine and construct an on-the-fly synchronised product with the MDP and the finite machine. The temporal structure acts as a guide for the RL agent within the product, where a modular Deep Deterministic Policy Gradient (DDPG) architecture is proposed to generate a low-level control policy. We evalua"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1909.11591","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/1909.11591/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":"1909.11591","created_at":"2026-07-05T00:21:37.178005+00:00"},{"alias_kind":"arxiv_version","alias_value":"1909.11591v2","created_at":"2026-07-05T00:21:37.178005+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1909.11591","created_at":"2026-07-05T00:21:37.178005+00:00"},{"alias_kind":"pith_short_12","alias_value":"3Z3GZJAPPJHN","created_at":"2026-07-05T00:21:37.178005+00:00"},{"alias_kind":"pith_short_16","alias_value":"3Z3GZJAPPJHNU3EF","created_at":"2026-07-05T00:21:37.178005+00:00"},{"alias_kind":"pith_short_8","alias_value":"3Z3GZJAP","created_at":"2026-07-05T00:21:37.178005+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.00442","citing_title":"Learning Gait-Aware Quadruped Locomotion with Temporal Logic Specifications","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24740","citing_title":"Reinforcement Learning for Reachability: Guaranteeing Asymptotic Optimality","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00838","citing_title":"Decoupled Behavioral Cloning for Scalable Inductive Generalization in RL from Specifications","ref_index":40,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3Z3GZJAPPJHNU3EFA3MN5AFEWV","json":"https://pith.science/pith/3Z3GZJAPPJHNU3EFA3MN5AFEWV.json","graph_json":"https://pith.science/api/pith-number/3Z3GZJAPPJHNU3EFA3MN5AFEWV/graph.json","events_json":"https://pith.science/api/pith-number/3Z3GZJAPPJHNU3EFA3MN5AFEWV/events.json","paper":"https://pith.science/paper/3Z3GZJAP"},"agent_actions":{"view_html":"https://pith.science/pith/3Z3GZJAPPJHNU3EFA3MN5AFEWV","download_json":"https://pith.science/pith/3Z3GZJAPPJHNU3EFA3MN5AFEWV.json","view_paper":"https://pith.science/paper/3Z3GZJAP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1909.11591&json=true","fetch_graph":"https://pith.science/api/pith-number/3Z3GZJAPPJHNU3EFA3MN5AFEWV/graph.json","fetch_events":"https://pith.science/api/pith-number/3Z3GZJAPPJHNU3EFA3MN5AFEWV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3Z3GZJAPPJHNU3EFA3MN5AFEWV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3Z3GZJAPPJHNU3EFA3MN5AFEWV/action/storage_attestation","attest_author":"https://pith.science/pith/3Z3GZJAPPJHNU3EFA3MN5AFEWV/action/author_attestation","sign_citation":"https://pith.science/pith/3Z3GZJAPPJHNU3EFA3MN5AFEWV/action/citation_signature","submit_replication":"https://pith.science/pith/3Z3GZJAPPJHNU3EFA3MN5AFEWV/action/replication_record"}},"created_at":"2026-07-05T00:21:37.178005+00:00","updated_at":"2026-07-05T00:21:37.178005+00:00"}