{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:I3VLMFEIC4TIDAJRPHPOPYXGAV","short_pith_number":"pith:I3VLMFEI","canonical_record":{"source":{"id":"2310.15418","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-10-24T00:22:19Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ffc311654196fc93f78f0dc4c19f6b0d6483717e85e86085f0ef9308a685be11","abstract_canon_sha256":"cfffe22fd2c3c7aa338f189ba0cb113def12efa7b3688387c2bb35868e952eb4"},"schema_version":"1.0"},"canonical_sha256":"46eab61488172681813179dee7e2e6054a2f19a0be5dbd659931ccd32f2acad5","source":{"kind":"arxiv","id":"2310.15418","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2310.15418","created_at":"2026-07-05T07:35:40Z"},{"alias_kind":"arxiv_version","alias_value":"2310.15418v1","created_at":"2026-07-05T07:35:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.15418","created_at":"2026-07-05T07:35:40Z"},{"alias_kind":"pith_short_12","alias_value":"I3VLMFEIC4TI","created_at":"2026-07-05T07:35:40Z"},{"alias_kind":"pith_short_16","alias_value":"I3VLMFEIC4TIDAJR","created_at":"2026-07-05T07:35:40Z"},{"alias_kind":"pith_short_8","alias_value":"I3VLMFEI","created_at":"2026-07-05T07:35:40Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:I3VLMFEIC4TIDAJRPHPOPYXGAV","target":"record","payload":{"canonical_record":{"source":{"id":"2310.15418","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-10-24T00:22:19Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ffc311654196fc93f78f0dc4c19f6b0d6483717e85e86085f0ef9308a685be11","abstract_canon_sha256":"cfffe22fd2c3c7aa338f189ba0cb113def12efa7b3688387c2bb35868e952eb4"},"schema_version":"1.0"},"canonical_sha256":"46eab61488172681813179dee7e2e6054a2f19a0be5dbd659931ccd32f2acad5","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:35:40.562852Z","signature_b64":"XGieC6mDI5kIBGO2ZZyT+PwJopgIegsIPGDOkkE3rxFsOz766moc112+Cn46DZxwEJxPRVNSq+m2Ww0tBzqsDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"46eab61488172681813179dee7e2e6054a2f19a0be5dbd659931ccd32f2acad5","last_reissued_at":"2026-07-05T07:35:40.562519Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:35:40.562519Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2310.15418","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:35:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"n8AS4WUXSSepkM5ZtlBANASoVHD7LvOSnuX8OT736xpax7E6UCh0Rbh0k8nTCK+z0o5/ky3p2tcBwycpQ+9VAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T01:30:57.470858Z"},"content_sha256":"d9533416e617c51cc412cb8f5787c1ebba11b75da9b1e3ca60131805e32da768","schema_version":"1.0","event_id":"sha256:d9533416e617c51cc412cb8f5787c1ebba11b75da9b1e3ca60131805e32da768"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:I3VLMFEIC4TIDAJRPHPOPYXGAV","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Fractal Landscapes in Policy Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Sicun Gao, Sylvia Herbert, Tao Wang","submitted_at":"2023-10-24T00:22:19Z","abstract_excerpt":"Policy gradient lies at the core of deep reinforcement learning (RL) in continuous domains. Despite much success, it is often observed in practice that RL training with policy gradient can fail for many reasons, even on standard control problems with known solutions. We propose a framework for understanding one inherent limitation of the policy gradient approach: the optimization landscape in the policy space can be extremely non-smooth or fractal for certain classes of MDPs, such that there does not exist gradient to be estimated in the first place. We draw on techniques from chaos theory and"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.15418","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/2310.15418/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":"2026-07-05T07:35:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Msw6d/SibkCrEzBbj/nq43CgpnrUKGxcNdbMDlEvT4JULCv5azUATc9ME6fbxU17j92pOjYz72IeVmmxhEZwBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T01:30:57.471374Z"},"content_sha256":"eb640b1e1bc84a5e7c5b4b5919d67d71a2e38621b2fd0721c10c8bddde4dd3f4","schema_version":"1.0","event_id":"sha256:eb640b1e1bc84a5e7c5b4b5919d67d71a2e38621b2fd0721c10c8bddde4dd3f4"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/I3VLMFEIC4TIDAJRPHPOPYXGAV/bundle.json","state_url":"https://pith.science/pith/I3VLMFEIC4TIDAJRPHPOPYXGAV/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/I3VLMFEIC4TIDAJRPHPOPYXGAV/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":"2026-08-08T01:30:57Z","links":{"resolver":"https://pith.science/pith/I3VLMFEIC4TIDAJRPHPOPYXGAV","bundle":"https://pith.science/pith/I3VLMFEIC4TIDAJRPHPOPYXGAV/bundle.json","state":"https://pith.science/pith/I3VLMFEIC4TIDAJRPHPOPYXGAV/state.json","well_known_bundle":"https://pith.science/.well-known/pith/I3VLMFEIC4TIDAJRPHPOPYXGAV/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:I3VLMFEIC4TIDAJRPHPOPYXGAV","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":"cfffe22fd2c3c7aa338f189ba0cb113def12efa7b3688387c2bb35868e952eb4","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-10-24T00:22:19Z","title_canon_sha256":"ffc311654196fc93f78f0dc4c19f6b0d6483717e85e86085f0ef9308a685be11"},"schema_version":"1.0","source":{"id":"2310.15418","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2310.15418","created_at":"2026-07-05T07:35:40Z"},{"alias_kind":"arxiv_version","alias_value":"2310.15418v1","created_at":"2026-07-05T07:35:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.15418","created_at":"2026-07-05T07:35:40Z"},{"alias_kind":"pith_short_12","alias_value":"I3VLMFEIC4TI","created_at":"2026-07-05T07:35:40Z"},{"alias_kind":"pith_short_16","alias_value":"I3VLMFEIC4TIDAJR","created_at":"2026-07-05T07:35:40Z"},{"alias_kind":"pith_short_8","alias_value":"I3VLMFEI","created_at":"2026-07-05T07:35:40Z"}],"graph_snapshots":[{"event_id":"sha256:eb640b1e1bc84a5e7c5b4b5919d67d71a2e38621b2fd0721c10c8bddde4dd3f4","target":"graph","created_at":"2026-07-05T07:35:40Z","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/2310.15418/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Policy gradient lies at the core of deep reinforcement learning (RL) in continuous domains. Despite much success, it is often observed in practice that RL training with policy gradient can fail for many reasons, even on standard control problems with known solutions. We propose a framework for understanding one inherent limitation of the policy gradient approach: the optimization landscape in the policy space can be extremely non-smooth or fractal for certain classes of MDPs, such that there does not exist gradient to be estimated in the first place. We draw on techniques from chaos theory and","authors_text":"Sicun Gao, Sylvia Herbert, Tao Wang","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-10-24T00:22:19Z","title":"Fractal Landscapes in Policy Optimization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.15418","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:d9533416e617c51cc412cb8f5787c1ebba11b75da9b1e3ca60131805e32da768","target":"record","created_at":"2026-07-05T07:35:40Z","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":"cfffe22fd2c3c7aa338f189ba0cb113def12efa7b3688387c2bb35868e952eb4","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-10-24T00:22:19Z","title_canon_sha256":"ffc311654196fc93f78f0dc4c19f6b0d6483717e85e86085f0ef9308a685be11"},"schema_version":"1.0","source":{"id":"2310.15418","kind":"arxiv","version":1}},"canonical_sha256":"46eab61488172681813179dee7e2e6054a2f19a0be5dbd659931ccd32f2acad5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"46eab61488172681813179dee7e2e6054a2f19a0be5dbd659931ccd32f2acad5","first_computed_at":"2026-07-05T07:35:40.562519Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:35:40.562519Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"XGieC6mDI5kIBGO2ZZyT+PwJopgIegsIPGDOkkE3rxFsOz766moc112+Cn46DZxwEJxPRVNSq+m2Ww0tBzqsDw==","signature_status":"signed_v1","signed_at":"2026-07-05T07:35:40.562852Z","signed_message":"canonical_sha256_bytes"},"source_id":"2310.15418","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d9533416e617c51cc412cb8f5787c1ebba11b75da9b1e3ca60131805e32da768","sha256:eb640b1e1bc84a5e7c5b4b5919d67d71a2e38621b2fd0721c10c8bddde4dd3f4"],"state_sha256":"23c8995c7714bef7bb48f249cc6ff5de43ca078732336080f6c4e436584dac62"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HnjIPBJgfUk3+hVcJwZRQ8p/+ctZrDpH+n97I25X6LqYIczh9c7y4tovW8TP1F4Vnqe11FF9zxkIshOZIF+QAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T01:30:57.476324Z","bundle_sha256":"c4478507430f52428a673646a1e11b4ffa9be6c4f1ba0eecce333ff34baf8712"}}