{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:SRHBKOBX5KJJXTVIALVE3D6MIA","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"2a8c7613c7d01d8297b16a0e594028b52eaf2f8f53e73b6ce66181d3f879a9b8","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-12-08T23:29:57Z","title_canon_sha256":"dfeff39f9b23a060712cf3330fda8ba481d99a81af6c5d38db40a9f6c17279b8"},"schema_version":"1.0","source":{"id":"2312.05405","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2312.05405","created_at":"2026-07-05T07:22:03Z"},{"alias_kind":"arxiv_version","alias_value":"2312.05405v1","created_at":"2026-07-05T07:22:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.05405","created_at":"2026-07-05T07:22:03Z"},{"alias_kind":"pith_short_12","alias_value":"SRHBKOBX5KJJ","created_at":"2026-07-05T07:22:03Z"},{"alias_kind":"pith_short_16","alias_value":"SRHBKOBX5KJJXTVI","created_at":"2026-07-05T07:22:03Z"},{"alias_kind":"pith_short_8","alias_value":"SRHBKOBX","created_at":"2026-07-05T07:22:03Z"}],"graph_snapshots":[{"event_id":"sha256:b6179d82d2b2e4cff07377bc815fb7fe60c5fc1d70649b604fb9cd16f8c5e98f","target":"graph","created_at":"2026-07-05T07:22:03Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2312.05405/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"On-policy reinforcement learning (RL) has become a popular framework for solving sequential decision problems due to its computational efficiency and theoretical simplicity. Some on-policy methods guarantee every policy update is constrained to a trust region relative to the prior policy to ensure training stability. These methods often require computationally intensive non-linear optimization or require a particular form of action distribution. In this work, we show that applying KL penalization alone is nearly sufficient to enforce such trust regions. Then, we show that introducing a \"fixup\"","authors_text":"Gaurav S. Sukhatme, K.R. Zentner, Ujjwal Puri, Zhehui Huang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-12-08T23:29:57Z","title":"Guaranteed Trust Region Optimization via Two-Phase KL Penalization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.05405","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:201dab7065217899df3c0ae1cf26de6bb50ad9810f0f1b6fedf05f3316b3014b","target":"record","created_at":"2026-07-05T07:22:03Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"2a8c7613c7d01d8297b16a0e594028b52eaf2f8f53e73b6ce66181d3f879a9b8","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-12-08T23:29:57Z","title_canon_sha256":"dfeff39f9b23a060712cf3330fda8ba481d99a81af6c5d38db40a9f6c17279b8"},"schema_version":"1.0","source":{"id":"2312.05405","kind":"arxiv","version":1}},"canonical_sha256":"944e153837ea929bcea802ea4d8fcc402f34168af0c506660c20ea5ea6890307","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"944e153837ea929bcea802ea4d8fcc402f34168af0c506660c20ea5ea6890307","first_computed_at":"2026-07-05T07:22:03.944301Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:22:03.944301Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"LtXt02oxMY1UkhebUvy3cIe0wkVY/Og7mvhay01OwC0k4y7mVRCT3/y6nYDh2JXvNE1+twxeIWjAkYscEUF1CQ==","signature_status":"signed_v1","signed_at":"2026-07-05T07:22:03.944761Z","signed_message":"canonical_sha256_bytes"},"source_id":"2312.05405","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:201dab7065217899df3c0ae1cf26de6bb50ad9810f0f1b6fedf05f3316b3014b","sha256:b6179d82d2b2e4cff07377bc815fb7fe60c5fc1d70649b604fb9cd16f8c5e98f"],"state_sha256":"72bb909d0d3a317427e0a166d9e4c6f2ec68f2794c0776530a46181856e61820"}