{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2017:CFKYWVYHVCUMSCO5IMJAX3W2LO","short_pith_number":"pith:CFKYWVYH","canonical_record":{"source":{"id":"1704.04651","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2017-04-15T15:38:23Z","cross_cats_sorted":[],"title_canon_sha256":"4c9d726b1a2108d46adf005994134ecfe8fcc60e08dd5199af1855b9e2db7bda","abstract_canon_sha256":"cca3cbd0158890019c41c410eb2b4d34a94f4ae2a0945835be264c24d450457c"},"schema_version":"1.0"},"canonical_sha256":"11558b5707a8a8c909dd43120beeda5ba5455946d447f627df34e0bbe39ea46c","source":{"kind":"arxiv","id":"1704.04651","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1704.04651","created_at":"2026-05-18T00:12:56Z"},{"alias_kind":"arxiv_version","alias_value":"1704.04651v2","created_at":"2026-05-18T00:12:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1704.04651","created_at":"2026-05-18T00:12:56Z"},{"alias_kind":"pith_short_12","alias_value":"CFKYWVYHVCUM","created_at":"2026-05-18T12:31:10Z"},{"alias_kind":"pith_short_16","alias_value":"CFKYWVYHVCUMSCO5","created_at":"2026-05-18T12:31:10Z"},{"alias_kind":"pith_short_8","alias_value":"CFKYWVYH","created_at":"2026-05-18T12:31:10Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2017:CFKYWVYHVCUMSCO5IMJAX3W2LO","target":"record","payload":{"canonical_record":{"source":{"id":"1704.04651","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2017-04-15T15:38:23Z","cross_cats_sorted":[],"title_canon_sha256":"4c9d726b1a2108d46adf005994134ecfe8fcc60e08dd5199af1855b9e2db7bda","abstract_canon_sha256":"cca3cbd0158890019c41c410eb2b4d34a94f4ae2a0945835be264c24d450457c"},"schema_version":"1.0"},"canonical_sha256":"11558b5707a8a8c909dd43120beeda5ba5455946d447f627df34e0bbe39ea46c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:12:56.951659Z","signature_b64":"mGbZMe6s73xqcqMsnXRlowhGakDy+kD5qeYyMDjBpzXGGvHWPnJTmRdWyYunjKprWlVEhfLlfXD1NWhg5k5RDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"11558b5707a8a8c909dd43120beeda5ba5455946d447f627df34e0bbe39ea46c","last_reissued_at":"2026-05-18T00:12:56.950919Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:12:56.950919Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1704.04651","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":"2026-05-18T00:12:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"bLkPvAvVWxcUHE+wFPr96mxi4k5jUa5pBVuJXOF7O81xje4YWFxB3VHPQcPU1O/R+dmnCC/rT8dJ4GqW01tLAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-11T20:43:47.200588Z"},"content_sha256":"6771572a8028ae8d3eeb8c9e48c2158eccd7ed098bebabf7098ba53bc8c74130","schema_version":"1.0","event_id":"sha256:6771572a8028ae8d3eeb8c9e48c2158eccd7ed098bebabf7098ba53bc8c74130"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2017:CFKYWVYHVCUMSCO5IMJAX3W2LO","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"The Reactor: A fast and sample-efficient Actor-Critic agent for Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Audrunas Gruslys, Bilal Piot, Marc Bellemare, Mohammad Gheshlaghi Azar, Remi Munos, Will Dabney","submitted_at":"2017-04-15T15:38:23Z","abstract_excerpt":"In this work we present a new agent architecture, called Reactor, which combines multiple algorithmic and architectural contributions to produce an agent with higher sample-efficiency than Prioritized Dueling DQN (Wang et al., 2016) and Categorical DQN (Bellemare et al., 2017), while giving better run-time performance than A3C (Mnih et al., 2016). Our first contribution is a new policy evaluation algorithm called Distributional Retrace, which brings multi-step off-policy updates to the distributional reinforcement learning setting. The same approach can be used to convert several classes of mu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1704.04651","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":""},"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":"2026-05-18T00:12:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vSAKaPeNOEs1WPa/erAwNetoRUBX4BSwmCi/ZwnQyQdbmJvN60ccbwbPzF18i/RXdCbJUScs9n+L8P56Ba/GDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-11T20:43:47.200976Z"},"content_sha256":"4f46b1008b990625a95dc1c12a2ec62e1e78b5e9a77c849a96f98f9237353738","schema_version":"1.0","event_id":"sha256:4f46b1008b990625a95dc1c12a2ec62e1e78b5e9a77c849a96f98f9237353738"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/CFKYWVYHVCUMSCO5IMJAX3W2LO/bundle.json","state_url":"https://pith.science/pith/CFKYWVYHVCUMSCO5IMJAX3W2LO/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/CFKYWVYHVCUMSCO5IMJAX3W2LO/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":"2026-06-11T20:43:47Z","links":{"resolver":"https://pith.science/pith/CFKYWVYHVCUMSCO5IMJAX3W2LO","bundle":"https://pith.science/pith/CFKYWVYHVCUMSCO5IMJAX3W2LO/bundle.json","state":"https://pith.science/pith/CFKYWVYHVCUMSCO5IMJAX3W2LO/state.json","well_known_bundle":"https://pith.science/.well-known/pith/CFKYWVYHVCUMSCO5IMJAX3W2LO/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2017:CFKYWVYHVCUMSCO5IMJAX3W2LO","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":"cca3cbd0158890019c41c410eb2b4d34a94f4ae2a0945835be264c24d450457c","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2017-04-15T15:38:23Z","title_canon_sha256":"4c9d726b1a2108d46adf005994134ecfe8fcc60e08dd5199af1855b9e2db7bda"},"schema_version":"1.0","source":{"id":"1704.04651","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1704.04651","created_at":"2026-05-18T00:12:56Z"},{"alias_kind":"arxiv_version","alias_value":"1704.04651v2","created_at":"2026-05-18T00:12:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1704.04651","created_at":"2026-05-18T00:12:56Z"},{"alias_kind":"pith_short_12","alias_value":"CFKYWVYHVCUM","created_at":"2026-05-18T12:31:10Z"},{"alias_kind":"pith_short_16","alias_value":"CFKYWVYHVCUMSCO5","created_at":"2026-05-18T12:31:10Z"},{"alias_kind":"pith_short_8","alias_value":"CFKYWVYH","created_at":"2026-05-18T12:31:10Z"}],"graph_snapshots":[{"event_id":"sha256:4f46b1008b990625a95dc1c12a2ec62e1e78b5e9a77c849a96f98f9237353738","target":"graph","created_at":"2026-05-18T00:12:56Z","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":"In this work we present a new agent architecture, called Reactor, which combines multiple algorithmic and architectural contributions to produce an agent with higher sample-efficiency than Prioritized Dueling DQN (Wang et al., 2016) and Categorical DQN (Bellemare et al., 2017), while giving better run-time performance than A3C (Mnih et al., 2016). Our first contribution is a new policy evaluation algorithm called Distributional Retrace, which brings multi-step off-policy updates to the distributional reinforcement learning setting. The same approach can be used to convert several classes of mu","authors_text":"Audrunas Gruslys, Bilal Piot, Marc Bellemare, Mohammad Gheshlaghi Azar, Remi Munos, Will Dabney","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2017-04-15T15:38:23Z","title":"The Reactor: A fast and sample-efficient Actor-Critic agent for Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1704.04651","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":"sha256:6771572a8028ae8d3eeb8c9e48c2158eccd7ed098bebabf7098ba53bc8c74130","target":"record","created_at":"2026-05-18T00:12:56Z","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":"cca3cbd0158890019c41c410eb2b4d34a94f4ae2a0945835be264c24d450457c","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2017-04-15T15:38:23Z","title_canon_sha256":"4c9d726b1a2108d46adf005994134ecfe8fcc60e08dd5199af1855b9e2db7bda"},"schema_version":"1.0","source":{"id":"1704.04651","kind":"arxiv","version":2}},"canonical_sha256":"11558b5707a8a8c909dd43120beeda5ba5455946d447f627df34e0bbe39ea46c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"11558b5707a8a8c909dd43120beeda5ba5455946d447f627df34e0bbe39ea46c","first_computed_at":"2026-05-18T00:12:56.950919Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:12:56.950919Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"mGbZMe6s73xqcqMsnXRlowhGakDy+kD5qeYyMDjBpzXGGvHWPnJTmRdWyYunjKprWlVEhfLlfXD1NWhg5k5RDA==","signature_status":"signed_v1","signed_at":"2026-05-18T00:12:56.951659Z","signed_message":"canonical_sha256_bytes"},"source_id":"1704.04651","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6771572a8028ae8d3eeb8c9e48c2158eccd7ed098bebabf7098ba53bc8c74130","sha256:4f46b1008b990625a95dc1c12a2ec62e1e78b5e9a77c849a96f98f9237353738"],"state_sha256":"72375f7d7c88389f883f7c6debbd6263c22d5efd5d06bd3771af330ee763e256"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FLQMYI7rEkohgEgSz8CI0Qj+KpflVVHHtPJAK8Ldb4YE9gv5tE/LWgja3IF/1YTzClgxCRP0INs0UX5Q3HOFAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-11T20:43:47.203689Z","bundle_sha256":"926dfa73f1b12b4b9c3e44bdc67d79f623ec108806ca527df8df0700d17bbe6f"}}