{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2016:7F6GGKJNFHZQHSIGUH5FM2DNKI","short_pith_number":"pith:7F6GGKJN","canonical_record":{"source":{"id":"1612.07548","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2016-12-22T11:30:35Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"6c6431045be9ca64a39da80a69e5ec5914622e6b67e63fb819c9df3b2979fe86","abstract_canon_sha256":"2f20450746050e63e5f8e7de68646202614b594f101fd233ee450fe14a999e4e"},"schema_version":"1.0"},"canonical_sha256":"f97c63292d29f303c906a1fa56686d523a58a9c4a5ef785cbed2915a8d7827ab","source":{"kind":"arxiv","id":"1612.07548","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1612.07548","created_at":"2026-05-18T00:54:09Z"},{"alias_kind":"arxiv_version","alias_value":"1612.07548v1","created_at":"2026-05-18T00:54:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1612.07548","created_at":"2026-05-18T00:54:09Z"},{"alias_kind":"pith_short_12","alias_value":"7F6GGKJNFHZQ","created_at":"2026-05-18T12:30:04Z"},{"alias_kind":"pith_short_16","alias_value":"7F6GGKJNFHZQHSIG","created_at":"2026-05-18T12:30:04Z"},{"alias_kind":"pith_short_8","alias_value":"7F6GGKJN","created_at":"2026-05-18T12:30:04Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2016:7F6GGKJNFHZQHSIGUH5FM2DNKI","target":"record","payload":{"canonical_record":{"source":{"id":"1612.07548","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2016-12-22T11:30:35Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"6c6431045be9ca64a39da80a69e5ec5914622e6b67e63fb819c9df3b2979fe86","abstract_canon_sha256":"2f20450746050e63e5f8e7de68646202614b594f101fd233ee450fe14a999e4e"},"schema_version":"1.0"},"canonical_sha256":"f97c63292d29f303c906a1fa56686d523a58a9c4a5ef785cbed2915a8d7827ab","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:54:09.786443Z","signature_b64":"5hvv26v25bDicXoOt1emNfC27XbMppdJOx+TET4DUgKv8+Mp7bpg9ijxRCuOJbu3R9heJJ2vFhDRRSP3Ip70AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f97c63292d29f303c906a1fa56686d523a58a9c4a5ef785cbed2915a8d7827ab","last_reissued_at":"2026-05-18T00:54:09.785975Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:54:09.785975Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1612.07548","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":"2026-05-18T00:54:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"a36kuO/wtXSZ/7njRf5fFElJYVoZ6K7m+X4opUM6KcXVhkEEwrtU0SJBvmFTdSHq8OQiSbHKR88TVOsjWXMzDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-06T15:36:44.694232Z"},"content_sha256":"2db12f38cadf5e96c4ccc8a5ddd9291c000770ce581fa10c74419f72351ec9bb","schema_version":"1.0","event_id":"sha256:2db12f38cadf5e96c4ccc8a5ddd9291c000770ce581fa10c74419f72351ec9bb"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2016:7F6GGKJNFHZQHSIGUH5FM2DNKI","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Non-Deterministic Policy Improvement Stabilizes Approximated Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cs.AI","authors_text":"Klaus Obermayer, Rong Guo, Wendelin B\\\"ohmer","submitted_at":"2016-12-22T11:30:35Z","abstract_excerpt":"This paper investigates a type of instability that is linked to the greedy policy improvement in approximated reinforcement learning. We show empirically that non-deterministic policy improvement can stabilize methods like LSPI by controlling the improvements' stochasticity. Additionally we show that a suitable representation of the value function also stabilizes the solution to some degree. The presented approach is simple and should also be easily transferable to more sophisticated algorithms like deep reinforcement learning."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1612.07548","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":""},"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":"2026-05-18T00:54:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PsUUTVw3P7PRWFD24cucEea6srgiGIuj/RDh2I7xFvwEn4Lkw5gcHuMQ1t+UNIIugRsKwJ5W3X0GCPVcV6IIDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-06T15:36:44.694930Z"},"content_sha256":"3107dd8413c81f1a4dab312ec6ed375988a08f0bff6499a483b24dd6a2ddc989","schema_version":"1.0","event_id":"sha256:3107dd8413c81f1a4dab312ec6ed375988a08f0bff6499a483b24dd6a2ddc989"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7F6GGKJNFHZQHSIGUH5FM2DNKI/bundle.json","state_url":"https://pith.science/pith/7F6GGKJNFHZQHSIGUH5FM2DNKI/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7F6GGKJNFHZQHSIGUH5FM2DNKI/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":"2026-06-06T15:36:44Z","links":{"resolver":"https://pith.science/pith/7F6GGKJNFHZQHSIGUH5FM2DNKI","bundle":"https://pith.science/pith/7F6GGKJNFHZQHSIGUH5FM2DNKI/bundle.json","state":"https://pith.science/pith/7F6GGKJNFHZQHSIGUH5FM2DNKI/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7F6GGKJNFHZQHSIGUH5FM2DNKI/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2016:7F6GGKJNFHZQHSIGUH5FM2DNKI","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":"2f20450746050e63e5f8e7de68646202614b594f101fd233ee450fe14a999e4e","cross_cats_sorted":["cs.LG","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2016-12-22T11:30:35Z","title_canon_sha256":"6c6431045be9ca64a39da80a69e5ec5914622e6b67e63fb819c9df3b2979fe86"},"schema_version":"1.0","source":{"id":"1612.07548","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1612.07548","created_at":"2026-05-18T00:54:09Z"},{"alias_kind":"arxiv_version","alias_value":"1612.07548v1","created_at":"2026-05-18T00:54:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1612.07548","created_at":"2026-05-18T00:54:09Z"},{"alias_kind":"pith_short_12","alias_value":"7F6GGKJNFHZQ","created_at":"2026-05-18T12:30:04Z"},{"alias_kind":"pith_short_16","alias_value":"7F6GGKJNFHZQHSIG","created_at":"2026-05-18T12:30:04Z"},{"alias_kind":"pith_short_8","alias_value":"7F6GGKJN","created_at":"2026-05-18T12:30:04Z"}],"graph_snapshots":[{"event_id":"sha256:3107dd8413c81f1a4dab312ec6ed375988a08f0bff6499a483b24dd6a2ddc989","target":"graph","created_at":"2026-05-18T00:54:09Z","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"},"paper":{"abstract_excerpt":"This paper investigates a type of instability that is linked to the greedy policy improvement in approximated reinforcement learning. We show empirically that non-deterministic policy improvement can stabilize methods like LSPI by controlling the improvements' stochasticity. Additionally we show that a suitable representation of the value function also stabilizes the solution to some degree. The presented approach is simple and should also be easily transferable to more sophisticated algorithms like deep reinforcement learning.","authors_text":"Klaus Obermayer, Rong Guo, Wendelin B\\\"ohmer","cross_cats":["cs.LG","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2016-12-22T11:30:35Z","title":"Non-Deterministic Policy Improvement Stabilizes Approximated Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1612.07548","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":"sha256:2db12f38cadf5e96c4ccc8a5ddd9291c000770ce581fa10c74419f72351ec9bb","target":"record","created_at":"2026-05-18T00:54:09Z","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":"2f20450746050e63e5f8e7de68646202614b594f101fd233ee450fe14a999e4e","cross_cats_sorted":["cs.LG","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2016-12-22T11:30:35Z","title_canon_sha256":"6c6431045be9ca64a39da80a69e5ec5914622e6b67e63fb819c9df3b2979fe86"},"schema_version":"1.0","source":{"id":"1612.07548","kind":"arxiv","version":1}},"canonical_sha256":"f97c63292d29f303c906a1fa56686d523a58a9c4a5ef785cbed2915a8d7827ab","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f97c63292d29f303c906a1fa56686d523a58a9c4a5ef785cbed2915a8d7827ab","first_computed_at":"2026-05-18T00:54:09.785975Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:54:09.785975Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"5hvv26v25bDicXoOt1emNfC27XbMppdJOx+TET4DUgKv8+Mp7bpg9ijxRCuOJbu3R9heJJ2vFhDRRSP3Ip70AQ==","signature_status":"signed_v1","signed_at":"2026-05-18T00:54:09.786443Z","signed_message":"canonical_sha256_bytes"},"source_id":"1612.07548","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2db12f38cadf5e96c4ccc8a5ddd9291c000770ce581fa10c74419f72351ec9bb","sha256:3107dd8413c81f1a4dab312ec6ed375988a08f0bff6499a483b24dd6a2ddc989"],"state_sha256":"3290eb5cd1d4aa467d7217b608fa5f96a711336913f0592c5c717deb39cca0fd"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GsfRBtG60gX96uYLLx/08TEDkm+Y8cwRPBzmtdR0Nixi/udj7gWfQz6MZVtO6I6HD1RhZyseDiR6yBXdc4atAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-06T15:36:44.698332Z","bundle_sha256":"54578b6963bf83d32a4419ee960a3d13ad106b2b3e46704902afba4a6fa353da"}}