{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:PWPJGPMQLFL3KC62S43IPD5RZY","short_pith_number":"pith:PWPJGPMQ","schema_version":"1.0","canonical_sha256":"7d9e933d905957b50bda9736878fb1ce2d72081aea003de2cc3f0096c80fa323","source":{"kind":"arxiv","id":"2309.06420","version":1},"attestation_state":"computed","paper":{"title":"Verifiable Reinforcement Learning Systems via Compositionality","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.SY"],"primary_cat":"eess.SY","authors_text":"Aryaman Singh Samyal, Christos Verginis, Cyrus Neary, Murat Cubuktepe, Ufuk Topcu","submitted_at":"2023-09-09T17:11:44Z","abstract_excerpt":"We propose a framework for verifiable and compositional reinforcement learning (RL) in which a collection of RL subsystems, each of which learns to accomplish a separate subtask, are composed to achieve an overall task. The framework consists of a high-level model, represented as a parametric Markov decision process, which is used to plan and analyze compositions of subsystems, and of the collection of low-level subsystems themselves. The subsystems are implemented as deep RL agents operating under partial observability. By defining interfaces between the subsystems, the framework enables auto"},"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":"2309.06420","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.SY","submitted_at":"2023-09-09T17:11:44Z","cross_cats_sorted":["cs.AI","cs.LG","cs.SY"],"title_canon_sha256":"3e91e5491cc48292cd9ce3b0d4797e24a51b42757abfd6d196cab8f1e15da067","abstract_canon_sha256":"f5e5ab9b49fde17a84517b0b5e9ca118509365e44ecb7e043ab0d37e3d402951"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:50:07.790592Z","signature_b64":"03W9HEW7KrD6+on7kUDG4o3pkc3I02QaU7XTwLAIBLf70ueONMJ2X58hx8TdMvsJoxCZrjb4aPWNJLoxiN+gAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7d9e933d905957b50bda9736878fb1ce2d72081aea003de2cc3f0096c80fa323","last_reissued_at":"2026-07-05T06:50:07.790061Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:50:07.790061Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Verifiable Reinforcement Learning Systems via Compositionality","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.SY"],"primary_cat":"eess.SY","authors_text":"Aryaman Singh Samyal, Christos Verginis, Cyrus Neary, Murat Cubuktepe, Ufuk Topcu","submitted_at":"2023-09-09T17:11:44Z","abstract_excerpt":"We propose a framework for verifiable and compositional reinforcement learning (RL) in which a collection of RL subsystems, each of which learns to accomplish a separate subtask, are composed to achieve an overall task. The framework consists of a high-level model, represented as a parametric Markov decision process, which is used to plan and analyze compositions of subsystems, and of the collection of low-level subsystems themselves. The subsystems are implemented as deep RL agents operating under partial observability. By defining interfaces between the subsystems, the framework enables auto"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.06420","kind":"arxiv","version":1},"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/2309.06420/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":"2309.06420","created_at":"2026-07-05T06:50:07.790125+00:00"},{"alias_kind":"arxiv_version","alias_value":"2309.06420v1","created_at":"2026-07-05T06:50:07.790125+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.06420","created_at":"2026-07-05T06:50:07.790125+00:00"},{"alias_kind":"pith_short_12","alias_value":"PWPJGPMQLFL3","created_at":"2026-07-05T06:50:07.790125+00:00"},{"alias_kind":"pith_short_16","alias_value":"PWPJGPMQLFL3KC62","created_at":"2026-07-05T06:50:07.790125+00:00"},{"alias_kind":"pith_short_8","alias_value":"PWPJGPMQ","created_at":"2026-07-05T06:50:07.790125+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.14001","citing_title":"VeRecycle: Reclaiming Guarantees from Probabilistic Certificates for Stochastic Dynamical Systems after Change","ref_index":29,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PWPJGPMQLFL3KC62S43IPD5RZY","json":"https://pith.science/pith/PWPJGPMQLFL3KC62S43IPD5RZY.json","graph_json":"https://pith.science/api/pith-number/PWPJGPMQLFL3KC62S43IPD5RZY/graph.json","events_json":"https://pith.science/api/pith-number/PWPJGPMQLFL3KC62S43IPD5RZY/events.json","paper":"https://pith.science/paper/PWPJGPMQ"},"agent_actions":{"view_html":"https://pith.science/pith/PWPJGPMQLFL3KC62S43IPD5RZY","download_json":"https://pith.science/pith/PWPJGPMQLFL3KC62S43IPD5RZY.json","view_paper":"https://pith.science/paper/PWPJGPMQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2309.06420&json=true","fetch_graph":"https://pith.science/api/pith-number/PWPJGPMQLFL3KC62S43IPD5RZY/graph.json","fetch_events":"https://pith.science/api/pith-number/PWPJGPMQLFL3KC62S43IPD5RZY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PWPJGPMQLFL3KC62S43IPD5RZY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PWPJGPMQLFL3KC62S43IPD5RZY/action/storage_attestation","attest_author":"https://pith.science/pith/PWPJGPMQLFL3KC62S43IPD5RZY/action/author_attestation","sign_citation":"https://pith.science/pith/PWPJGPMQLFL3KC62S43IPD5RZY/action/citation_signature","submit_replication":"https://pith.science/pith/PWPJGPMQLFL3KC62S43IPD5RZY/action/replication_record"}},"created_at":"2026-07-05T06:50:07.790125+00:00","updated_at":"2026-07-05T06:50:07.790125+00:00"}