{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:GVK7FQMTWVGAUAFFQ536IG7H6Y","short_pith_number":"pith:GVK7FQMT","schema_version":"1.0","canonical_sha256":"3555f2c193b54c0a00a58777e41be7f62eefc923e51c7892fd53ca82946730c6","source":{"kind":"arxiv","id":"2603.18396","version":3},"attestation_state":"computed","paper":{"title":"RE-SAC: Disentangling aleatoric and epistemic risks in bus fleet control: A stable and robust ensemble DRL approach","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.RO"],"primary_cat":"cs.LG","authors_text":"Liang Zheng, Yifan Zhang","submitted_at":"2026-03-19T01:37:45Z","abstract_excerpt":"Bus holding control is challenging due to stochastic traffic and passenger demand. While deep reinforcement learning (DRL) shows promise, standard actor-critic algorithms suffer from Q-value instability in volatile environments. A key source of this instability is the conflation of two distinct uncertainties: aleatoric uncertainty (irreducible noise) and epistemic uncertainty (data insufficiency). Treating these as a single risk leads to value underestimation in noisy states, causing catastrophic policy collapse. We propose a robust ensemble soft actor-critic (RE-SAC) framework to explicitly d"},"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":"2603.18396","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-03-19T01:37:45Z","cross_cats_sorted":["cs.RO"],"title_canon_sha256":"c07f4e40c5bc8b4333323751815b805531b8859a5c7a34ad39a84f0c1590397a","abstract_canon_sha256":"464935306dc9b6a5446bea42f5d773f1da8583a974f2d9c7bc8285ce60de0523"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:01:40.252571Z","signature_b64":"7KDK+cwUy7Fijm0O3wtIlEr6N8+lgH60Ws8mY2wKbMljAhoYUZDQMt2uOY8isK9IQFaNxnEm/2QCeiux8uwZDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3555f2c193b54c0a00a58777e41be7f62eefc923e51c7892fd53ca82946730c6","last_reissued_at":"2026-05-20T00:01:40.251996Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:01:40.251996Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RE-SAC: Disentangling aleatoric and epistemic risks in bus fleet control: A stable and robust ensemble DRL approach","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.RO"],"primary_cat":"cs.LG","authors_text":"Liang Zheng, Yifan Zhang","submitted_at":"2026-03-19T01:37:45Z","abstract_excerpt":"Bus holding control is challenging due to stochastic traffic and passenger demand. While deep reinforcement learning (DRL) shows promise, standard actor-critic algorithms suffer from Q-value instability in volatile environments. A key source of this instability is the conflation of two distinct uncertainties: aleatoric uncertainty (irreducible noise) and epistemic uncertainty (data insufficiency). Treating these as a single risk leads to value underestimation in noisy states, causing catastrophic policy collapse. We propose a robust ensemble soft actor-critic (RE-SAC) framework to explicitly d"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.18396","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/2603.18396/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":"2603.18396","created_at":"2026-05-20T00:01:40.252099+00:00"},{"alias_kind":"arxiv_version","alias_value":"2603.18396v3","created_at":"2026-05-20T00:01:40.252099+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.18396","created_at":"2026-05-20T00:01:40.252099+00:00"},{"alias_kind":"pith_short_12","alias_value":"GVK7FQMTWVGA","created_at":"2026-05-20T00:01:40.252099+00:00"},{"alias_kind":"pith_short_16","alias_value":"GVK7FQMTWVGAUAFF","created_at":"2026-05-20T00:01:40.252099+00:00"},{"alias_kind":"pith_short_8","alias_value":"GVK7FQMT","created_at":"2026-05-20T00:01:40.252099+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.16170","citing_title":"BAPR: Bayesian amnesic piecewise-robust reinforcement learning for non-stationary continuous control","ref_index":24,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GVK7FQMTWVGAUAFFQ536IG7H6Y","json":"https://pith.science/pith/GVK7FQMTWVGAUAFFQ536IG7H6Y.json","graph_json":"https://pith.science/api/pith-number/GVK7FQMTWVGAUAFFQ536IG7H6Y/graph.json","events_json":"https://pith.science/api/pith-number/GVK7FQMTWVGAUAFFQ536IG7H6Y/events.json","paper":"https://pith.science/paper/GVK7FQMT"},"agent_actions":{"view_html":"https://pith.science/pith/GVK7FQMTWVGAUAFFQ536IG7H6Y","download_json":"https://pith.science/pith/GVK7FQMTWVGAUAFFQ536IG7H6Y.json","view_paper":"https://pith.science/paper/GVK7FQMT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2603.18396&json=true","fetch_graph":"https://pith.science/api/pith-number/GVK7FQMTWVGAUAFFQ536IG7H6Y/graph.json","fetch_events":"https://pith.science/api/pith-number/GVK7FQMTWVGAUAFFQ536IG7H6Y/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GVK7FQMTWVGAUAFFQ536IG7H6Y/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GVK7FQMTWVGAUAFFQ536IG7H6Y/action/storage_attestation","attest_author":"https://pith.science/pith/GVK7FQMTWVGAUAFFQ536IG7H6Y/action/author_attestation","sign_citation":"https://pith.science/pith/GVK7FQMTWVGAUAFFQ536IG7H6Y/action/citation_signature","submit_replication":"https://pith.science/pith/GVK7FQMTWVGAUAFFQ536IG7H6Y/action/replication_record"}},"created_at":"2026-05-20T00:01:40.252099+00:00","updated_at":"2026-05-20T00:01:40.252099+00:00"}