{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:JVHZQ3AKQL5RTMXTFB5NQS4ZJT","short_pith_number":"pith:JVHZQ3AK","schema_version":"1.0","canonical_sha256":"4d4f986c0a82fb19b2f3287ad84b994cf05b11e1a902ddeb413115b2d650a77e","source":{"kind":"arxiv","id":"2606.10228","version":1},"attestation_state":"computed","paper":{"title":"SHAPO: Sharpness-Aware Policy Optimization for Safe Exploration","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.RO"],"primary_cat":"cs.LG","authors_text":"Kaustubh Mani, Liam Paull, Vincent Mai, Yann Pequignot","submitted_at":"2026-06-08T22:40:45Z","abstract_excerpt":"Safe exploration is a prerequisite for deploying reinforcement learning (RL) agents in safety-critical domains. In this paper, we approach safe exploration through the lens of epistemic uncertainty, where the actor's sensitivity to parameter perturbations serves as a practical proxy for regions of high uncertainty. We propose Sharpness-Aware Policy Optimization (SHAPO), a sharpness-aware policy update rule that evaluates gradients at perturbed parameters, making policy updates pessimistic with respect to the actor's epistemic uncertainty. Analytically we show that this adjustment implicitly re"},"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":"2606.10228","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-08T22:40:45Z","cross_cats_sorted":["cs.AI","cs.RO"],"title_canon_sha256":"b98e99029f29f96c4ae5e4e4c8db53cb8950754abafeb6dd66dd805f417e56af","abstract_canon_sha256":"b9966b5caf65b48669b9b8fc236f5c200e27978aa08539040c17e9a750af265b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-10T01:09:00.628993Z","signature_b64":"mBbnP0blFzCenv6GEb3Rl/4PZiHwGZAJYEIfeYPa6PkJJ2llbtAQBHG55eW5REtPfdHMnslUR5Mra+Wgj4QRAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4d4f986c0a82fb19b2f3287ad84b994cf05b11e1a902ddeb413115b2d650a77e","last_reissued_at":"2026-06-10T01:09:00.628155Z","signature_status":"signed_v1","first_computed_at":"2026-06-10T01:09:00.628155Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SHAPO: Sharpness-Aware Policy Optimization for Safe Exploration","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.RO"],"primary_cat":"cs.LG","authors_text":"Kaustubh Mani, Liam Paull, Vincent Mai, Yann Pequignot","submitted_at":"2026-06-08T22:40:45Z","abstract_excerpt":"Safe exploration is a prerequisite for deploying reinforcement learning (RL) agents in safety-critical domains. In this paper, we approach safe exploration through the lens of epistemic uncertainty, where the actor's sensitivity to parameter perturbations serves as a practical proxy for regions of high uncertainty. We propose Sharpness-Aware Policy Optimization (SHAPO), a sharpness-aware policy update rule that evaluates gradients at perturbed parameters, making policy updates pessimistic with respect to the actor's epistemic uncertainty. Analytically we show that this adjustment implicitly re"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.10228","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/2606.10228/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":"2606.10228","created_at":"2026-06-10T01:09:00.628280+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.10228v1","created_at":"2026-06-10T01:09:00.628280+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.10228","created_at":"2026-06-10T01:09:00.628280+00:00"},{"alias_kind":"pith_short_12","alias_value":"JVHZQ3AKQL5R","created_at":"2026-06-10T01:09:00.628280+00:00"},{"alias_kind":"pith_short_16","alias_value":"JVHZQ3AKQL5RTMXT","created_at":"2026-06-10T01:09:00.628280+00:00"},{"alias_kind":"pith_short_8","alias_value":"JVHZQ3AK","created_at":"2026-06-10T01:09:00.628280+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/JVHZQ3AKQL5RTMXTFB5NQS4ZJT","json":"https://pith.science/pith/JVHZQ3AKQL5RTMXTFB5NQS4ZJT.json","graph_json":"https://pith.science/api/pith-number/JVHZQ3AKQL5RTMXTFB5NQS4ZJT/graph.json","events_json":"https://pith.science/api/pith-number/JVHZQ3AKQL5RTMXTFB5NQS4ZJT/events.json","paper":"https://pith.science/paper/JVHZQ3AK"},"agent_actions":{"view_html":"https://pith.science/pith/JVHZQ3AKQL5RTMXTFB5NQS4ZJT","download_json":"https://pith.science/pith/JVHZQ3AKQL5RTMXTFB5NQS4ZJT.json","view_paper":"https://pith.science/paper/JVHZQ3AK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.10228&json=true","fetch_graph":"https://pith.science/api/pith-number/JVHZQ3AKQL5RTMXTFB5NQS4ZJT/graph.json","fetch_events":"https://pith.science/api/pith-number/JVHZQ3AKQL5RTMXTFB5NQS4ZJT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JVHZQ3AKQL5RTMXTFB5NQS4ZJT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JVHZQ3AKQL5RTMXTFB5NQS4ZJT/action/storage_attestation","attest_author":"https://pith.science/pith/JVHZQ3AKQL5RTMXTFB5NQS4ZJT/action/author_attestation","sign_citation":"https://pith.science/pith/JVHZQ3AKQL5RTMXTFB5NQS4ZJT/action/citation_signature","submit_replication":"https://pith.science/pith/JVHZQ3AKQL5RTMXTFB5NQS4ZJT/action/replication_record"}},"created_at":"2026-06-10T01:09:00.628280+00:00","updated_at":"2026-06-10T01:09:00.628280+00:00"}