{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:ON5XCBDNPSHYSCUYJ5KIKDCNJR","short_pith_number":"pith:ON5XCBDN","canonical_record":{"source":{"id":"2110.11280","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-10-21T17:06:59Z","cross_cats_sorted":[],"title_canon_sha256":"cf1ed7a46d8eba0ac6cbcd24f34a88dd47952fdad1f103c595d92e7ba6b50eff","abstract_canon_sha256":"b1c0b3696c85b61e06441b2a89a2ad7789b369d2281a340837ca9e73de2ee1e2"},"schema_version":"1.0"},"canonical_sha256":"737b71046d7c8f890a984f54850c4d4c76e402ede5e3ca6ce8809046831fd4e9","source":{"kind":"arxiv","id":"2110.11280","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2110.11280","created_at":"2026-07-05T04:04:42Z"},{"alias_kind":"arxiv_version","alias_value":"2110.11280v2","created_at":"2026-07-05T04:04:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.11280","created_at":"2026-07-05T04:04:42Z"},{"alias_kind":"pith_short_12","alias_value":"ON5XCBDNPSHY","created_at":"2026-07-05T04:04:42Z"},{"alias_kind":"pith_short_16","alias_value":"ON5XCBDNPSHYSCUY","created_at":"2026-07-05T04:04:42Z"},{"alias_kind":"pith_short_8","alias_value":"ON5XCBDN","created_at":"2026-07-05T04:04:42Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:ON5XCBDNPSHYSCUYJ5KIKDCNJR","target":"record","payload":{"canonical_record":{"source":{"id":"2110.11280","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-10-21T17:06:59Z","cross_cats_sorted":[],"title_canon_sha256":"cf1ed7a46d8eba0ac6cbcd24f34a88dd47952fdad1f103c595d92e7ba6b50eff","abstract_canon_sha256":"b1c0b3696c85b61e06441b2a89a2ad7789b369d2281a340837ca9e73de2ee1e2"},"schema_version":"1.0"},"canonical_sha256":"737b71046d7c8f890a984f54850c4d4c76e402ede5e3ca6ce8809046831fd4e9","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:04:42.896526Z","signature_b64":"G+vkzrqMW/x4rbZk8/Bar+FtI+SjzGAijfnG8JIw4qsYBxQFH/sLy1NkJ9T1dQQ5RnI2FvAcysCnitzVVUmmAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"737b71046d7c8f890a984f54850c4d4c76e402ede5e3ca6ce8809046831fd4e9","last_reissued_at":"2026-07-05T04:04:42.896084Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:04:42.896084Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2110.11280","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T04:04:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OsqYNTpombtgp7NVaK5PWz4uwpcQ6jiGQ8YEr6WX5lJtfAnpimHZbK+P9FwZ1ZSx4hGyog4O/pvMLvRj2vkKAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T04:06:59.902736Z"},"content_sha256":"547f07c28f8c0caffb9ccd8f047374c3bdc6468513d1d425845c3044f003cbdc","schema_version":"1.0","event_id":"sha256:547f07c28f8c0caffb9ccd8f047374c3bdc6468513d1d425845c3044f003cbdc"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:ON5XCBDNPSHYSCUYJ5KIKDCNJR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Actor-critic is implicitly biased towards high entropy optimal policies","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Matus Telgarsky, Yuzheng Hu, Ziwei Ji","submitted_at":"2021-10-21T17:06:59Z","abstract_excerpt":"We show that the simplest actor-critic method -- a linear softmax policy updated with TD through interaction with a linear MDP, but featuring no explicit regularization or exploration -- does not merely find an optimal policy, but moreover prefers high entropy optimal policies. To demonstrate the strength of this bias, the algorithm not only has no regularization, no projections, and no exploration like $\\epsilon$-greedy, but is moreover trained on a single trajectory with no resets. The key consequence of the high entropy bias is that uniform mixing assumptions on the MDP, which exist in some"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.11280","kind":"arxiv","version":2},"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/2110.11280/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":"2026-07-05T04:04:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ZM+sgJ9jFnCJliy283bZidZ8R5mHzf0fljoYg1IjHHgSR2cAefDf7YWV/UgcNcfCXdPMjB2WzLfF7jMGhnvTBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T04:06:59.903204Z"},"content_sha256":"c647f4b62f27714c1a59cdae03850d0cadf5c0ddfc6dce32640fedbcf1bdb14e","schema_version":"1.0","event_id":"sha256:c647f4b62f27714c1a59cdae03850d0cadf5c0ddfc6dce32640fedbcf1bdb14e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ON5XCBDNPSHYSCUYJ5KIKDCNJR/bundle.json","state_url":"https://pith.science/pith/ON5XCBDNPSHYSCUYJ5KIKDCNJR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ON5XCBDNPSHYSCUYJ5KIKDCNJR/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":"2026-08-06T04:06:59Z","links":{"resolver":"https://pith.science/pith/ON5XCBDNPSHYSCUYJ5KIKDCNJR","bundle":"https://pith.science/pith/ON5XCBDNPSHYSCUYJ5KIKDCNJR/bundle.json","state":"https://pith.science/pith/ON5XCBDNPSHYSCUYJ5KIKDCNJR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ON5XCBDNPSHYSCUYJ5KIKDCNJR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:ON5XCBDNPSHYSCUYJ5KIKDCNJR","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":"b1c0b3696c85b61e06441b2a89a2ad7789b369d2281a340837ca9e73de2ee1e2","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-10-21T17:06:59Z","title_canon_sha256":"cf1ed7a46d8eba0ac6cbcd24f34a88dd47952fdad1f103c595d92e7ba6b50eff"},"schema_version":"1.0","source":{"id":"2110.11280","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2110.11280","created_at":"2026-07-05T04:04:42Z"},{"alias_kind":"arxiv_version","alias_value":"2110.11280v2","created_at":"2026-07-05T04:04:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.11280","created_at":"2026-07-05T04:04:42Z"},{"alias_kind":"pith_short_12","alias_value":"ON5XCBDNPSHY","created_at":"2026-07-05T04:04:42Z"},{"alias_kind":"pith_short_16","alias_value":"ON5XCBDNPSHYSCUY","created_at":"2026-07-05T04:04:42Z"},{"alias_kind":"pith_short_8","alias_value":"ON5XCBDN","created_at":"2026-07-05T04:04:42Z"}],"graph_snapshots":[{"event_id":"sha256:c647f4b62f27714c1a59cdae03850d0cadf5c0ddfc6dce32640fedbcf1bdb14e","target":"graph","created_at":"2026-07-05T04:04:42Z","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/2110.11280/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We show that the simplest actor-critic method -- a linear softmax policy updated with TD through interaction with a linear MDP, but featuring no explicit regularization or exploration -- does not merely find an optimal policy, but moreover prefers high entropy optimal policies. To demonstrate the strength of this bias, the algorithm not only has no regularization, no projections, and no exploration like $\\epsilon$-greedy, but is moreover trained on a single trajectory with no resets. The key consequence of the high entropy bias is that uniform mixing assumptions on the MDP, which exist in some","authors_text":"Matus Telgarsky, Yuzheng Hu, Ziwei Ji","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-10-21T17:06:59Z","title":"Actor-critic is implicitly biased towards high entropy optimal policies"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.11280","kind":"arxiv","version":2},"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":"sha256:547f07c28f8c0caffb9ccd8f047374c3bdc6468513d1d425845c3044f003cbdc","target":"record","created_at":"2026-07-05T04:04:42Z","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":"b1c0b3696c85b61e06441b2a89a2ad7789b369d2281a340837ca9e73de2ee1e2","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-10-21T17:06:59Z","title_canon_sha256":"cf1ed7a46d8eba0ac6cbcd24f34a88dd47952fdad1f103c595d92e7ba6b50eff"},"schema_version":"1.0","source":{"id":"2110.11280","kind":"arxiv","version":2}},"canonical_sha256":"737b71046d7c8f890a984f54850c4d4c76e402ede5e3ca6ce8809046831fd4e9","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"737b71046d7c8f890a984f54850c4d4c76e402ede5e3ca6ce8809046831fd4e9","first_computed_at":"2026-07-05T04:04:42.896084Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T04:04:42.896084Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"G+vkzrqMW/x4rbZk8/Bar+FtI+SjzGAijfnG8JIw4qsYBxQFH/sLy1NkJ9T1dQQ5RnI2FvAcysCnitzVVUmmAw==","signature_status":"signed_v1","signed_at":"2026-07-05T04:04:42.896526Z","signed_message":"canonical_sha256_bytes"},"source_id":"2110.11280","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:547f07c28f8c0caffb9ccd8f047374c3bdc6468513d1d425845c3044f003cbdc","sha256:c647f4b62f27714c1a59cdae03850d0cadf5c0ddfc6dce32640fedbcf1bdb14e"],"state_sha256":"e7b537b0fa899e5b63e6a00223ae9a8ad461d4a7d4da77d6d4b31314ddf9b65e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LuD6gUzbLYwYnUN9JJd0yoQAw7KeqXuKWGGnh8Qde5978FgLvX9omI1iKlrOGngzn42l82TuH/OA715/8cLMBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T04:06:59.906473Z","bundle_sha256":"9259d86948cd900b987d6d48611b9b7ee97632e4e098df630bbfea6ea3101fe5"}}