{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:SU6AG5QDLQRXA5P4SLDYU6Z4LV","short_pith_number":"pith:SU6AG5QD","canonical_record":{"source":{"id":"2508.04225","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-08-06T09:01:29Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"73df0777c4da1c0ba5e7c20a2731602b62c7403c58bd3c0930462e2e249e3bcc","abstract_canon_sha256":"211990872c08e9e7ae70760e742fb816c2eb6ee08831739f406b4c03341c596a"},"schema_version":"1.0"},"canonical_sha256":"953c0376035c237075fc92c78a7b3c5d6f5f93da65bd06a46b2cc0c04c1f7792","source":{"kind":"arxiv","id":"2508.04225","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.04225","created_at":"2026-07-21T00:19:50Z"},{"alias_kind":"arxiv_version","alias_value":"2508.04225v4","created_at":"2026-07-21T00:19:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.04225","created_at":"2026-07-21T00:19:50Z"},{"alias_kind":"pith_short_12","alias_value":"SU6AG5QDLQRX","created_at":"2026-07-21T00:19:50Z"},{"alias_kind":"pith_short_16","alias_value":"SU6AG5QDLQRXA5P4","created_at":"2026-07-21T00:19:50Z"},{"alias_kind":"pith_short_8","alias_value":"SU6AG5QD","created_at":"2026-07-21T00:19:50Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:SU6AG5QDLQRXA5P4SLDYU6Z4LV","target":"record","payload":{"canonical_record":{"source":{"id":"2508.04225","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-08-06T09:01:29Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"73df0777c4da1c0ba5e7c20a2731602b62c7403c58bd3c0930462e2e249e3bcc","abstract_canon_sha256":"211990872c08e9e7ae70760e742fb816c2eb6ee08831739f406b4c03341c596a"},"schema_version":"1.0"},"canonical_sha256":"953c0376035c237075fc92c78a7b3c5d6f5f93da65bd06a46b2cc0c04c1f7792","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-21T00:19:50.395809Z","signature_b64":"34brWiusROzXG8hlEFipnSE/tN9uJYD8kxFjApKOUz/P9U8Jrit3eVtTssK/spwkfBZTCRKSEk2z0BiHIbkNBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"953c0376035c237075fc92c78a7b3c5d6f5f93da65bd06a46b2cc0c04c1f7792","last_reissued_at":"2026-07-21T00:19:50.394810Z","signature_status":"signed_v1","first_computed_at":"2026-07-21T00:19:50.394810Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2508.04225","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-21T00:19:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DVdsit4oVbowgRP+4/4RblcplphPIdZmcEyTLP6NqpF7FXMPXA0feuUoV6MwyM703P1dUt82iaIEDaGx2plGDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T23:37:31.281892Z"},"content_sha256":"5caca627446df800f4a60ef499184c615f2609763d8966b1360c2fc933ba6645","schema_version":"1.0","event_id":"sha256:5caca627446df800f4a60ef499184c615f2609763d8966b1360c2fc933ba6645"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:SU6AG5QDLQRXA5P4SLDYU6Z4LV","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Symmetric Behavior Regularized Policy Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Haseeb Shah, Lingwei Zhu, Martha White, Zheng Chen","submitted_at":"2025-08-06T09:01:29Z","abstract_excerpt":"Behavior Regularized Policy Optimization (BRPO) leverages asymmetric divergence regularization to mitigate distribution shift in offline reinforcement learning. This paper is the first to study the open question of symmetric BRPO. Using didactic examples, we show that symmetric regularization can outperform asymmetric regularization in addressing one-sided bias, near-boundary policy updates, and projection geometry consistency. However, symmetric divergences do not fit BRPO naturally: they do not permit a closed-form solution when used as regularizers, and can lead to numerical instability whe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.04225","kind":"arxiv","version":4},"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/2508.04225/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-21T00:19:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AStmBXdGNxttjQAdl0E4V3AJh7An2sbLJQU+fIi+pEMb5nYgV56rzQRh/UaSJXAmgCB2xVGCjpbw+XTFqXeHAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T23:37:31.282660Z"},"content_sha256":"cbe5eb108c42aecfdeec4a8135640eb2f0d94390a74167dfb0f59fe7f628d7e2","schema_version":"1.0","event_id":"sha256:cbe5eb108c42aecfdeec4a8135640eb2f0d94390a74167dfb0f59fe7f628d7e2"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SU6AG5QDLQRXA5P4SLDYU6Z4LV/bundle.json","state_url":"https://pith.science/pith/SU6AG5QDLQRXA5P4SLDYU6Z4LV/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SU6AG5QDLQRXA5P4SLDYU6Z4LV/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-16T23:37:31Z","links":{"resolver":"https://pith.science/pith/SU6AG5QDLQRXA5P4SLDYU6Z4LV","bundle":"https://pith.science/pith/SU6AG5QDLQRXA5P4SLDYU6Z4LV/bundle.json","state":"https://pith.science/pith/SU6AG5QDLQRXA5P4SLDYU6Z4LV/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SU6AG5QDLQRXA5P4SLDYU6Z4LV/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:SU6AG5QDLQRXA5P4SLDYU6Z4LV","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":"211990872c08e9e7ae70760e742fb816c2eb6ee08831739f406b4c03341c596a","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-08-06T09:01:29Z","title_canon_sha256":"73df0777c4da1c0ba5e7c20a2731602b62c7403c58bd3c0930462e2e249e3bcc"},"schema_version":"1.0","source":{"id":"2508.04225","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.04225","created_at":"2026-07-21T00:19:50Z"},{"alias_kind":"arxiv_version","alias_value":"2508.04225v4","created_at":"2026-07-21T00:19:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.04225","created_at":"2026-07-21T00:19:50Z"},{"alias_kind":"pith_short_12","alias_value":"SU6AG5QDLQRX","created_at":"2026-07-21T00:19:50Z"},{"alias_kind":"pith_short_16","alias_value":"SU6AG5QDLQRXA5P4","created_at":"2026-07-21T00:19:50Z"},{"alias_kind":"pith_short_8","alias_value":"SU6AG5QD","created_at":"2026-07-21T00:19:50Z"}],"graph_snapshots":[{"event_id":"sha256:cbe5eb108c42aecfdeec4a8135640eb2f0d94390a74167dfb0f59fe7f628d7e2","target":"graph","created_at":"2026-07-21T00:19:50Z","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/2508.04225/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Behavior Regularized Policy Optimization (BRPO) leverages asymmetric divergence regularization to mitigate distribution shift in offline reinforcement learning. This paper is the first to study the open question of symmetric BRPO. Using didactic examples, we show that symmetric regularization can outperform asymmetric regularization in addressing one-sided bias, near-boundary policy updates, and projection geometry consistency. However, symmetric divergences do not fit BRPO naturally: they do not permit a closed-form solution when used as regularizers, and can lead to numerical instability whe","authors_text":"Haseeb Shah, Lingwei Zhu, Martha White, Zheng Chen","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-08-06T09:01:29Z","title":"Symmetric Behavior Regularized Policy Optimization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.04225","kind":"arxiv","version":4},"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":"sha256:5caca627446df800f4a60ef499184c615f2609763d8966b1360c2fc933ba6645","target":"record","created_at":"2026-07-21T00:19:50Z","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":"211990872c08e9e7ae70760e742fb816c2eb6ee08831739f406b4c03341c596a","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-08-06T09:01:29Z","title_canon_sha256":"73df0777c4da1c0ba5e7c20a2731602b62c7403c58bd3c0930462e2e249e3bcc"},"schema_version":"1.0","source":{"id":"2508.04225","kind":"arxiv","version":4}},"canonical_sha256":"953c0376035c237075fc92c78a7b3c5d6f5f93da65bd06a46b2cc0c04c1f7792","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"953c0376035c237075fc92c78a7b3c5d6f5f93da65bd06a46b2cc0c04c1f7792","first_computed_at":"2026-07-21T00:19:50.394810Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-21T00:19:50.394810Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"34brWiusROzXG8hlEFipnSE/tN9uJYD8kxFjApKOUz/P9U8Jrit3eVtTssK/spwkfBZTCRKSEk2z0BiHIbkNBw==","signature_status":"signed_v1","signed_at":"2026-07-21T00:19:50.395809Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.04225","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5caca627446df800f4a60ef499184c615f2609763d8966b1360c2fc933ba6645","sha256:cbe5eb108c42aecfdeec4a8135640eb2f0d94390a74167dfb0f59fe7f628d7e2"],"state_sha256":"9929c8b430bfb6e533d82dea2d05b59a5139079b12ca8162481340ba93bd37eb"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kWSAXLyLtyRS28YF5lbQc358M9oJIKq6mVfFV+cpIEEFbplqBWIUWzeA9gZF5RxrE2fgKyNsgu5t3CMu6bN6Bw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-16T23:37:31.287961Z","bundle_sha256":"3469e85de44712d5b5f4d4bc396d8e04f75d8d190fbdeca31cb9a82357807694"}}