{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2020:K5T3MAIRYQWZP5FCEFLCXDBIZ6","short_pith_number":"pith:K5T3MAIR","canonical_record":{"source":{"id":"2010.05380","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-10-12T00:50:17Z","cross_cats_sorted":[],"title_canon_sha256":"67fab7c535b64ff8f649d8a14bcb8ff004995317583317be6487c61ed1aa3477","abstract_canon_sha256":"175349d0a7bcbd2a2c3113ae524f2d4d5e3936daa463c541da2c1d12fe1b2088"},"schema_version":"1.0"},"canonical_sha256":"5767b60111c42d97f4a221562b8c28cf8c39d575153c8ff9393ae67333dd446a","source":{"kind":"arxiv","id":"2010.05380","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2010.05380","created_at":"2026-07-05T02:24:11Z"},{"alias_kind":"arxiv_version","alias_value":"2010.05380v4","created_at":"2026-07-05T02:24:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.05380","created_at":"2026-07-05T02:24:11Z"},{"alias_kind":"pith_short_12","alias_value":"K5T3MAIRYQWZ","created_at":"2026-07-05T02:24:11Z"},{"alias_kind":"pith_short_16","alias_value":"K5T3MAIRYQWZP5FC","created_at":"2026-07-05T02:24:11Z"},{"alias_kind":"pith_short_8","alias_value":"K5T3MAIR","created_at":"2026-07-05T02:24:11Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2020:K5T3MAIRYQWZP5FCEFLCXDBIZ6","target":"record","payload":{"canonical_record":{"source":{"id":"2010.05380","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-10-12T00:50:17Z","cross_cats_sorted":[],"title_canon_sha256":"67fab7c535b64ff8f649d8a14bcb8ff004995317583317be6487c61ed1aa3477","abstract_canon_sha256":"175349d0a7bcbd2a2c3113ae524f2d4d5e3936daa463c541da2c1d12fe1b2088"},"schema_version":"1.0"},"canonical_sha256":"5767b60111c42d97f4a221562b8c28cf8c39d575153c8ff9393ae67333dd446a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:24:11.391226Z","signature_b64":"p/yxkdico084sYMjMtQgZxlJT2T1CsrNuN7prH6/9QqNXiu+FbgdGl3nwFQiA8BVND3/vQetgVt/toqH38fCAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5767b60111c42d97f4a221562b8c28cf8c39d575153c8ff9393ae67333dd446a","last_reissued_at":"2026-07-05T02:24:11.390739Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:24:11.390739Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2010.05380","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":"2026-07-05T02:24:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pZwKNaZhEA/0iU3mdYHxAgkb/8Cczt29vHoM5gH+ndGJBFLMroJIIwE2Z4XBLQL7sOgIG5dzlWZlD4MW6JL4Dw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T16:01:30.075920Z"},"content_sha256":"e6f9abd23f36f1096ac1a6347278e5c559f6829925a978f7102bbfe8053afe50","schema_version":"1.0","event_id":"sha256:e6f9abd23f36f1096ac1a6347278e5c559f6829925a978f7102bbfe8053afe50"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2020:K5T3MAIRYQWZP5FCEFLCXDBIZ6","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Efficient Wasserstein Natural Gradients for Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Arthur Gretton, Ferenc Huszar, Michael Arbel, Ted Moskovitz","submitted_at":"2020-10-12T00:50:17Z","abstract_excerpt":"A novel optimization approach is proposed for application to policy gradient methods and evolution strategies for reinforcement learning (RL). The procedure uses a computationally efficient Wasserstein natural gradient (WNG) descent that takes advantage of the geometry induced by a Wasserstein penalty to speed optimization. This method follows the recent theme in RL of including a divergence penalty in the objective to establish a trust region. Experiments on challenging tasks demonstrate improvements in both computational cost and performance over advanced baselines."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.05380","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/2010.05380/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":"2026-07-05T02:24:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wShYXlAz4YA+L8MFJCsPXi/g/pedB9+822zgkonjrH+LqTN9lIz0FhSIRum4PzyYZ+rUnE1O5FetUbw7NQbQBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T16:01:30.076742Z"},"content_sha256":"fd3fb7b949b99794621bb76b206205b59b3a83c5480ab0ed62845fed98fa55c2","schema_version":"1.0","event_id":"sha256:fd3fb7b949b99794621bb76b206205b59b3a83c5480ab0ed62845fed98fa55c2"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/K5T3MAIRYQWZP5FCEFLCXDBIZ6/bundle.json","state_url":"https://pith.science/pith/K5T3MAIRYQWZP5FCEFLCXDBIZ6/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/K5T3MAIRYQWZP5FCEFLCXDBIZ6/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":"2026-08-17T16:01:30Z","links":{"resolver":"https://pith.science/pith/K5T3MAIRYQWZP5FCEFLCXDBIZ6","bundle":"https://pith.science/pith/K5T3MAIRYQWZP5FCEFLCXDBIZ6/bundle.json","state":"https://pith.science/pith/K5T3MAIRYQWZP5FCEFLCXDBIZ6/state.json","well_known_bundle":"https://pith.science/.well-known/pith/K5T3MAIRYQWZP5FCEFLCXDBIZ6/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:K5T3MAIRYQWZP5FCEFLCXDBIZ6","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":"175349d0a7bcbd2a2c3113ae524f2d4d5e3936daa463c541da2c1d12fe1b2088","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-10-12T00:50:17Z","title_canon_sha256":"67fab7c535b64ff8f649d8a14bcb8ff004995317583317be6487c61ed1aa3477"},"schema_version":"1.0","source":{"id":"2010.05380","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2010.05380","created_at":"2026-07-05T02:24:11Z"},{"alias_kind":"arxiv_version","alias_value":"2010.05380v4","created_at":"2026-07-05T02:24:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.05380","created_at":"2026-07-05T02:24:11Z"},{"alias_kind":"pith_short_12","alias_value":"K5T3MAIRYQWZ","created_at":"2026-07-05T02:24:11Z"},{"alias_kind":"pith_short_16","alias_value":"K5T3MAIRYQWZP5FC","created_at":"2026-07-05T02:24:11Z"},{"alias_kind":"pith_short_8","alias_value":"K5T3MAIR","created_at":"2026-07-05T02:24:11Z"}],"graph_snapshots":[{"event_id":"sha256:fd3fb7b949b99794621bb76b206205b59b3a83c5480ab0ed62845fed98fa55c2","target":"graph","created_at":"2026-07-05T02:24:11Z","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/2010.05380/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"A novel optimization approach is proposed for application to policy gradient methods and evolution strategies for reinforcement learning (RL). The procedure uses a computationally efficient Wasserstein natural gradient (WNG) descent that takes advantage of the geometry induced by a Wasserstein penalty to speed optimization. This method follows the recent theme in RL of including a divergence penalty in the objective to establish a trust region. Experiments on challenging tasks demonstrate improvements in both computational cost and performance over advanced baselines.","authors_text":"Arthur Gretton, Ferenc Huszar, Michael Arbel, Ted Moskovitz","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-10-12T00:50:17Z","title":"Efficient Wasserstein Natural Gradients for Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.05380","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:e6f9abd23f36f1096ac1a6347278e5c559f6829925a978f7102bbfe8053afe50","target":"record","created_at":"2026-07-05T02:24:11Z","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":"175349d0a7bcbd2a2c3113ae524f2d4d5e3936daa463c541da2c1d12fe1b2088","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-10-12T00:50:17Z","title_canon_sha256":"67fab7c535b64ff8f649d8a14bcb8ff004995317583317be6487c61ed1aa3477"},"schema_version":"1.0","source":{"id":"2010.05380","kind":"arxiv","version":4}},"canonical_sha256":"5767b60111c42d97f4a221562b8c28cf8c39d575153c8ff9393ae67333dd446a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5767b60111c42d97f4a221562b8c28cf8c39d575153c8ff9393ae67333dd446a","first_computed_at":"2026-07-05T02:24:11.390739Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T02:24:11.390739Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"p/yxkdico084sYMjMtQgZxlJT2T1CsrNuN7prH6/9QqNXiu+FbgdGl3nwFQiA8BVND3/vQetgVt/toqH38fCAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T02:24:11.391226Z","signed_message":"canonical_sha256_bytes"},"source_id":"2010.05380","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e6f9abd23f36f1096ac1a6347278e5c559f6829925a978f7102bbfe8053afe50","sha256:fd3fb7b949b99794621bb76b206205b59b3a83c5480ab0ed62845fed98fa55c2"],"state_sha256":"30dc9b862c96b46af7aa70106513154dbcddb33f5d7825c4c143b64e3c334423"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"QyWFX9ddLO5gqf83BmsidAIEUHDaY3YKEpq4OZayDIgS+stN5yUzrHbmJmHLnjOwUT2cSEwKIibJ+qqI29vlCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-17T16:01:30.079397Z","bundle_sha256":"4c43d40e6053b469bbfc0d142b99756ffa2c5bbb6d2d42acaeece25565041470"}}