{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:2BSHMVRF36SEHRUYENCDGAXWRC","short_pith_number":"pith:2BSHMVRF","schema_version":"1.0","canonical_sha256":"d064765625dfa443c69823443302f6888dac70c2316c68e34f24a760507955fe","source":{"kind":"arxiv","id":"2312.06436","version":2},"attestation_state":"computed","paper":{"title":"Reward Certification for Policy Smoothed Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Leandro Soriano Marcolino, Ronghui Mu, Tianle Zhang, Wenjie Ruan, Xiaowei Huang, Yanghao Zhang","submitted_at":"2023-12-11T15:07:58Z","abstract_excerpt":"Reinforcement Learning (RL) has achieved remarkable success in safety-critical areas, but it can be weakened by adversarial attacks. Recent studies have introduced \"smoothed policies\" in order to enhance its robustness. Yet, it is still challenging to establish a provable guarantee to certify the bound of its total reward. Prior methods relied primarily on computing bounds using Lipschitz continuity or calculating the probability of cumulative reward above specific thresholds. However, these techniques are only suited for continuous perturbations on the RL agent's observations and are restrict"},"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":"2312.06436","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-12-11T15:07:58Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b748f709f8081041239c5a0a2f71b6566f3e36f542b8bcd400ab532ba036e983","abstract_canon_sha256":"61b3734babf55eccfee7fbe45cd15f56270db83d0c75b9af4d2929a2416bc46a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:23:11.246147Z","signature_b64":"tOD0wxccrDM4E+QlUDID76y3rzo90hx1FAZh20COIyUbQB9YINeQZ7LT2apmkTsyrqz4yS/tGkk8a7eM8nyJDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d064765625dfa443c69823443302f6888dac70c2316c68e34f24a760507955fe","last_reissued_at":"2026-07-05T07:23:11.245701Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:23:11.245701Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reward Certification for Policy Smoothed Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Leandro Soriano Marcolino, Ronghui Mu, Tianle Zhang, Wenjie Ruan, Xiaowei Huang, Yanghao Zhang","submitted_at":"2023-12-11T15:07:58Z","abstract_excerpt":"Reinforcement Learning (RL) has achieved remarkable success in safety-critical areas, but it can be weakened by adversarial attacks. Recent studies have introduced \"smoothed policies\" in order to enhance its robustness. Yet, it is still challenging to establish a provable guarantee to certify the bound of its total reward. Prior methods relied primarily on computing bounds using Lipschitz continuity or calculating the probability of cumulative reward above specific thresholds. However, these techniques are only suited for continuous perturbations on the RL agent's observations and are restrict"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.06436","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/2312.06436/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":"2312.06436","created_at":"2026-07-05T07:23:11.245760+00:00"},{"alias_kind":"arxiv_version","alias_value":"2312.06436v2","created_at":"2026-07-05T07:23:11.245760+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.06436","created_at":"2026-07-05T07:23:11.245760+00:00"},{"alias_kind":"pith_short_12","alias_value":"2BSHMVRF36SE","created_at":"2026-07-05T07:23:11.245760+00:00"},{"alias_kind":"pith_short_16","alias_value":"2BSHMVRF36SEHRUY","created_at":"2026-07-05T07:23:11.245760+00:00"},{"alias_kind":"pith_short_8","alias_value":"2BSHMVRF","created_at":"2026-07-05T07:23:11.245760+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2BSHMVRF36SEHRUYENCDGAXWRC","json":"https://pith.science/pith/2BSHMVRF36SEHRUYENCDGAXWRC.json","graph_json":"https://pith.science/api/pith-number/2BSHMVRF36SEHRUYENCDGAXWRC/graph.json","events_json":"https://pith.science/api/pith-number/2BSHMVRF36SEHRUYENCDGAXWRC/events.json","paper":"https://pith.science/paper/2BSHMVRF"},"agent_actions":{"view_html":"https://pith.science/pith/2BSHMVRF36SEHRUYENCDGAXWRC","download_json":"https://pith.science/pith/2BSHMVRF36SEHRUYENCDGAXWRC.json","view_paper":"https://pith.science/paper/2BSHMVRF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2312.06436&json=true","fetch_graph":"https://pith.science/api/pith-number/2BSHMVRF36SEHRUYENCDGAXWRC/graph.json","fetch_events":"https://pith.science/api/pith-number/2BSHMVRF36SEHRUYENCDGAXWRC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2BSHMVRF36SEHRUYENCDGAXWRC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2BSHMVRF36SEHRUYENCDGAXWRC/action/storage_attestation","attest_author":"https://pith.science/pith/2BSHMVRF36SEHRUYENCDGAXWRC/action/author_attestation","sign_citation":"https://pith.science/pith/2BSHMVRF36SEHRUYENCDGAXWRC/action/citation_signature","submit_replication":"https://pith.science/pith/2BSHMVRF36SEHRUYENCDGAXWRC/action/replication_record"}},"created_at":"2026-07-05T07:23:11.245760+00:00","updated_at":"2026-07-05T07:23:11.245760+00:00"}