{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:ADY4X6ARGB7T42NAT5LO32OAUB","short_pith_number":"pith:ADY4X6AR","canonical_record":{"source":{"id":"1812.02648","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2018-12-06T16:36:20Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"aa47f17602f9cbdae55044b4a5ca8d5a0c631d2b03fdfbacf5e29b73391d078e","abstract_canon_sha256":"0580d25c5a5eb04c82d2db5958a7e1206197b628130588ad2d329d0cb2d39963"},"schema_version":"1.0"},"canonical_sha256":"00f1cbf811307f3e69a09f56ede9c0a04d020ccef314e837cc1aaf737f35cfb5","source":{"kind":"arxiv","id":"1812.02648","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1812.02648","created_at":"2026-05-17T23:58:55Z"},{"alias_kind":"arxiv_version","alias_value":"1812.02648v1","created_at":"2026-05-17T23:58:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1812.02648","created_at":"2026-05-17T23:58:55Z"},{"alias_kind":"pith_short_12","alias_value":"ADY4X6ARGB7T","created_at":"2026-05-18T12:32:13Z"},{"alias_kind":"pith_short_16","alias_value":"ADY4X6ARGB7T42NA","created_at":"2026-05-18T12:32:13Z"},{"alias_kind":"pith_short_8","alias_value":"ADY4X6AR","created_at":"2026-05-18T12:32:13Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:ADY4X6ARGB7T42NAT5LO32OAUB","target":"record","payload":{"canonical_record":{"source":{"id":"1812.02648","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2018-12-06T16:36:20Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"aa47f17602f9cbdae55044b4a5ca8d5a0c631d2b03fdfbacf5e29b73391d078e","abstract_canon_sha256":"0580d25c5a5eb04c82d2db5958a7e1206197b628130588ad2d329d0cb2d39963"},"schema_version":"1.0"},"canonical_sha256":"00f1cbf811307f3e69a09f56ede9c0a04d020ccef314e837cc1aaf737f35cfb5","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:58:55.256462Z","signature_b64":"DwPr1SKbvvHxBURz6TMfImQMujGdwIEg3cnuZWC3b5wEbZB/Trww3AEEkzLPxEJ2BTV1L/ANa597VlGE7tLRBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"00f1cbf811307f3e69a09f56ede9c0a04d020ccef314e837cc1aaf737f35cfb5","last_reissued_at":"2026-05-17T23:58:55.256070Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:58:55.256070Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1812.02648","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":"2026-05-17T23:58:55Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cL1HjXSjhRvZGneD3vB9gr3EjqewJcWjuQlgHMpZD0G3y1tqwjSD+UTrSvuza/u65SldpRm/dvAllE+NQX3vAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T01:32:45.823450Z"},"content_sha256":"bc89ce5498674fb2b7440be90345022bb275648c58793bbb9902b58f9316f67d","schema_version":"1.0","event_id":"sha256:bc89ce5498674fb2b7440be90345022bb275648c58793bbb9902b58f9316f67d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:ADY4X6ARGB7T42NAT5LO32OAUB","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Deep Reinforcement Learning and the Deadly Triad","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Florian Strub, Hado van Hasselt, Joseph Modayil, Matteo Hessel, Nicolas Sonnerat, Yotam Doron","submitted_at":"2018-12-06T16:36:20Z","abstract_excerpt":"We know from reinforcement learning theory that temporal difference learning can fail in certain cases. Sutton and Barto (2018) identify a deadly triad of function approximation, bootstrapping, and off-policy learning. When these three properties are combined, learning can diverge with the value estimates becoming unbounded. However, several algorithms successfully combine these three properties, which indicates that there is at least a partial gap in our understanding. In this work, we investigate the impact of the deadly triad in practice, in the context of a family of popular deep reinforce"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1812.02648","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":"2026-05-17T23:58:55Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gzAcCos/U4sUeIOspt8iaNCzQow621la4cMMFVaPLVd8rRDlqsP4iCmeQQvvJlvb5LjCN3wTASUzYI1Hf/lLBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T01:32:45.824151Z"},"content_sha256":"18d8db18018d07a3f80a13ede1c7f5f4122870dcf9c9046a9bb6e7e31aee1fcb","schema_version":"1.0","event_id":"sha256:18d8db18018d07a3f80a13ede1c7f5f4122870dcf9c9046a9bb6e7e31aee1fcb"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ADY4X6ARGB7T42NAT5LO32OAUB/bundle.json","state_url":"https://pith.science/pith/ADY4X6ARGB7T42NAT5LO32OAUB/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ADY4X6ARGB7T42NAT5LO32OAUB/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":"2026-05-27T01:32:45Z","links":{"resolver":"https://pith.science/pith/ADY4X6ARGB7T42NAT5LO32OAUB","bundle":"https://pith.science/pith/ADY4X6ARGB7T42NAT5LO32OAUB/bundle.json","state":"https://pith.science/pith/ADY4X6ARGB7T42NAT5LO32OAUB/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ADY4X6ARGB7T42NAT5LO32OAUB/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:ADY4X6ARGB7T42NAT5LO32OAUB","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":"0580d25c5a5eb04c82d2db5958a7e1206197b628130588ad2d329d0cb2d39963","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2018-12-06T16:36:20Z","title_canon_sha256":"aa47f17602f9cbdae55044b4a5ca8d5a0c631d2b03fdfbacf5e29b73391d078e"},"schema_version":"1.0","source":{"id":"1812.02648","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1812.02648","created_at":"2026-05-17T23:58:55Z"},{"alias_kind":"arxiv_version","alias_value":"1812.02648v1","created_at":"2026-05-17T23:58:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1812.02648","created_at":"2026-05-17T23:58:55Z"},{"alias_kind":"pith_short_12","alias_value":"ADY4X6ARGB7T","created_at":"2026-05-18T12:32:13Z"},{"alias_kind":"pith_short_16","alias_value":"ADY4X6ARGB7T42NA","created_at":"2026-05-18T12:32:13Z"},{"alias_kind":"pith_short_8","alias_value":"ADY4X6AR","created_at":"2026-05-18T12:32:13Z"}],"graph_snapshots":[{"event_id":"sha256:18d8db18018d07a3f80a13ede1c7f5f4122870dcf9c9046a9bb6e7e31aee1fcb","target":"graph","created_at":"2026-05-17T23:58:55Z","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":"We know from reinforcement learning theory that temporal difference learning can fail in certain cases. Sutton and Barto (2018) identify a deadly triad of function approximation, bootstrapping, and off-policy learning. When these three properties are combined, learning can diverge with the value estimates becoming unbounded. However, several algorithms successfully combine these three properties, which indicates that there is at least a partial gap in our understanding. In this work, we investigate the impact of the deadly triad in practice, in the context of a family of popular deep reinforce","authors_text":"Florian Strub, Hado van Hasselt, Joseph Modayil, Matteo Hessel, Nicolas Sonnerat, Yotam Doron","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2018-12-06T16:36:20Z","title":"Deep Reinforcement Learning and the Deadly Triad"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1812.02648","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:bc89ce5498674fb2b7440be90345022bb275648c58793bbb9902b58f9316f67d","target":"record","created_at":"2026-05-17T23:58:55Z","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":"0580d25c5a5eb04c82d2db5958a7e1206197b628130588ad2d329d0cb2d39963","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2018-12-06T16:36:20Z","title_canon_sha256":"aa47f17602f9cbdae55044b4a5ca8d5a0c631d2b03fdfbacf5e29b73391d078e"},"schema_version":"1.0","source":{"id":"1812.02648","kind":"arxiv","version":1}},"canonical_sha256":"00f1cbf811307f3e69a09f56ede9c0a04d020ccef314e837cc1aaf737f35cfb5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"00f1cbf811307f3e69a09f56ede9c0a04d020ccef314e837cc1aaf737f35cfb5","first_computed_at":"2026-05-17T23:58:55.256070Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:58:55.256070Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"DwPr1SKbvvHxBURz6TMfImQMujGdwIEg3cnuZWC3b5wEbZB/Trww3AEEkzLPxEJ2BTV1L/ANa597VlGE7tLRBQ==","signature_status":"signed_v1","signed_at":"2026-05-17T23:58:55.256462Z","signed_message":"canonical_sha256_bytes"},"source_id":"1812.02648","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:bc89ce5498674fb2b7440be90345022bb275648c58793bbb9902b58f9316f67d","sha256:18d8db18018d07a3f80a13ede1c7f5f4122870dcf9c9046a9bb6e7e31aee1fcb"],"state_sha256":"11651ad105da56ae5e3101909d410b1bdac38429b59067bd8b9161b7644583c6"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kPN2nC2XYKwqYa4Vl2J5JpvaeacPBLyJNxVySrnC9p6erbUvkZjxa3at6/2V7DWOeyu0p2ka/IcjjLvicHVTDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-27T01:32:45.827631Z","bundle_sha256":"d744c30daed3b54ff902086054e01b9e8d16d2c1770994ff29931394f1fc7c2e"}}