{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:4NWABOF5UHTJQOMVQNN6RVVWB3","short_pith_number":"pith:4NWABOF5","canonical_record":{"source":{"id":"1903.08894","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-03-21T09:42:41Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2cb9ccea37a2aeda9425c2ce2279ab7d1999ef891a4550d7b54b8d250ce8b675","abstract_canon_sha256":"893a883be8f7977e05faffce176fd26bb6af2d744dcd6397e74544ecebee37ee"},"schema_version":"1.0"},"canonical_sha256":"e36c00b8bda1e6983995835be8d6b60ed9937553f8ea205d628c42621c67d0cf","source":{"kind":"arxiv","id":"1903.08894","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1903.08894","created_at":"2026-05-17T23:50:43Z"},{"alias_kind":"arxiv_version","alias_value":"1903.08894v1","created_at":"2026-05-17T23:50:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1903.08894","created_at":"2026-05-17T23:50:43Z"},{"alias_kind":"pith_short_12","alias_value":"4NWABOF5UHTJ","created_at":"2026-05-18T12:33:10Z"},{"alias_kind":"pith_short_16","alias_value":"4NWABOF5UHTJQOMV","created_at":"2026-05-18T12:33:10Z"},{"alias_kind":"pith_short_8","alias_value":"4NWABOF5","created_at":"2026-05-18T12:33:10Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:4NWABOF5UHTJQOMVQNN6RVVWB3","target":"record","payload":{"canonical_record":{"source":{"id":"1903.08894","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-03-21T09:42:41Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2cb9ccea37a2aeda9425c2ce2279ab7d1999ef891a4550d7b54b8d250ce8b675","abstract_canon_sha256":"893a883be8f7977e05faffce176fd26bb6af2d744dcd6397e74544ecebee37ee"},"schema_version":"1.0"},"canonical_sha256":"e36c00b8bda1e6983995835be8d6b60ed9937553f8ea205d628c42621c67d0cf","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:50:43.647121Z","signature_b64":"ETGLZRyip5iyyP9XU2ZF0vHJ3dA15AbvdlEEqzTPpdX1MHpHuRzQpnHf+gAFNdZnJVa6SbY0+2go7imgW75IDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e36c00b8bda1e6983995835be8d6b60ed9937553f8ea205d628c42621c67d0cf","last_reissued_at":"2026-05-17T23:50:43.646611Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:50:43.646611Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1903.08894","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":"2026-05-17T23:50:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fbSZDAYjHtCufQVlhkVnWnMJXYC+C73QKaglC/GptpGre1kbvDfKn+WVWg6lLnoOPTjVEz4lFFSDmcY9OQv9Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T08:05:23.965592Z"},"content_sha256":"318baa3b7925427d0241c7c9dff3f244e4944fdba12ba7a984e7bc880132df29","schema_version":"1.0","event_id":"sha256:318baa3b7925427d0241c7c9dff3f244e4944fdba12ba7a984e7bc880132df29"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:4NWABOF5UHTJQOMVQNN6RVVWB3","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Towards Characterizing Divergence in Deep Q-Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Ethan Knight, Joshua Achiam, Pieter Abbeel","submitted_at":"2019-03-21T09:42:41Z","abstract_excerpt":"Deep Q-Learning (DQL), a family of temporal difference algorithms for control, employs three techniques collectively known as the `deadly triad' in reinforcement learning: bootstrapping, off-policy learning, and function approximation. Prior work has demonstrated that together these can lead to divergence in Q-learning algorithms, but the conditions under which divergence occurs are not well-understood. In this note, we give a simple analysis based on a linear approximation to the Q-value updates, which we believe provides insight into divergence under the deadly triad. The central point in ou"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1903.08894","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":"2026-05-17T23:50:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kS7UCWETqNneARrJEN+RmcgeWPGRBn1nmu/3apmyLVBWv350auIUxXzUljl/CVAdoBPkcUrPY6atYaTLnAEfDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T08:05:23.966034Z"},"content_sha256":"7e188c5bdd90e1bcc2663704d5ead0f17b47a4a0ddf90deff852c585ccfe1ea1","schema_version":"1.0","event_id":"sha256:7e188c5bdd90e1bcc2663704d5ead0f17b47a4a0ddf90deff852c585ccfe1ea1"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4NWABOF5UHTJQOMVQNN6RVVWB3/bundle.json","state_url":"https://pith.science/pith/4NWABOF5UHTJQOMVQNN6RVVWB3/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4NWABOF5UHTJQOMVQNN6RVVWB3/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":"2026-05-27T08:05:23Z","links":{"resolver":"https://pith.science/pith/4NWABOF5UHTJQOMVQNN6RVVWB3","bundle":"https://pith.science/pith/4NWABOF5UHTJQOMVQNN6RVVWB3/bundle.json","state":"https://pith.science/pith/4NWABOF5UHTJQOMVQNN6RVVWB3/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4NWABOF5UHTJQOMVQNN6RVVWB3/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:4NWABOF5UHTJQOMVQNN6RVVWB3","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":"893a883be8f7977e05faffce176fd26bb6af2d744dcd6397e74544ecebee37ee","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-03-21T09:42:41Z","title_canon_sha256":"2cb9ccea37a2aeda9425c2ce2279ab7d1999ef891a4550d7b54b8d250ce8b675"},"schema_version":"1.0","source":{"id":"1903.08894","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1903.08894","created_at":"2026-05-17T23:50:43Z"},{"alias_kind":"arxiv_version","alias_value":"1903.08894v1","created_at":"2026-05-17T23:50:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1903.08894","created_at":"2026-05-17T23:50:43Z"},{"alias_kind":"pith_short_12","alias_value":"4NWABOF5UHTJ","created_at":"2026-05-18T12:33:10Z"},{"alias_kind":"pith_short_16","alias_value":"4NWABOF5UHTJQOMV","created_at":"2026-05-18T12:33:10Z"},{"alias_kind":"pith_short_8","alias_value":"4NWABOF5","created_at":"2026-05-18T12:33:10Z"}],"graph_snapshots":[{"event_id":"sha256:7e188c5bdd90e1bcc2663704d5ead0f17b47a4a0ddf90deff852c585ccfe1ea1","target":"graph","created_at":"2026-05-17T23:50:43Z","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":"Deep Q-Learning (DQL), a family of temporal difference algorithms for control, employs three techniques collectively known as the `deadly triad' in reinforcement learning: bootstrapping, off-policy learning, and function approximation. Prior work has demonstrated that together these can lead to divergence in Q-learning algorithms, but the conditions under which divergence occurs are not well-understood. In this note, we give a simple analysis based on a linear approximation to the Q-value updates, which we believe provides insight into divergence under the deadly triad. The central point in ou","authors_text":"Ethan Knight, Joshua Achiam, Pieter Abbeel","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-03-21T09:42:41Z","title":"Towards Characterizing Divergence in Deep Q-Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1903.08894","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":"sha256:318baa3b7925427d0241c7c9dff3f244e4944fdba12ba7a984e7bc880132df29","target":"record","created_at":"2026-05-17T23:50:43Z","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":"893a883be8f7977e05faffce176fd26bb6af2d744dcd6397e74544ecebee37ee","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-03-21T09:42:41Z","title_canon_sha256":"2cb9ccea37a2aeda9425c2ce2279ab7d1999ef891a4550d7b54b8d250ce8b675"},"schema_version":"1.0","source":{"id":"1903.08894","kind":"arxiv","version":1}},"canonical_sha256":"e36c00b8bda1e6983995835be8d6b60ed9937553f8ea205d628c42621c67d0cf","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e36c00b8bda1e6983995835be8d6b60ed9937553f8ea205d628c42621c67d0cf","first_computed_at":"2026-05-17T23:50:43.646611Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:50:43.646611Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ETGLZRyip5iyyP9XU2ZF0vHJ3dA15AbvdlEEqzTPpdX1MHpHuRzQpnHf+gAFNdZnJVa6SbY0+2go7imgW75IDA==","signature_status":"signed_v1","signed_at":"2026-05-17T23:50:43.647121Z","signed_message":"canonical_sha256_bytes"},"source_id":"1903.08894","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:318baa3b7925427d0241c7c9dff3f244e4944fdba12ba7a984e7bc880132df29","sha256:7e188c5bdd90e1bcc2663704d5ead0f17b47a4a0ddf90deff852c585ccfe1ea1"],"state_sha256":"2c0478a29df0c44df1536ea91ed58fb406310400f759c32e59a06456564aaf42"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Ruf8SHBSQpzTqauQGR0mOrK2z9I8OZUdvZak5kCMcHAyxG1Dz301w0MpAASk53bftQ3OPhjzv6wxiIrQ813gBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-27T08:05:23.969499Z","bundle_sha256":"c958ba9c20880b584c1a81413d60e27ff4139d1a0d3974b157522d2449771f56"}}