{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:7ZR66RUXQ3VQE6XJ5DS6IC75A5","short_pith_number":"pith:7ZR66RUX","canonical_record":{"source":{"id":"1806.10230","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NE","submitted_at":"2018-06-26T22:14:36Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"5d87f51ac364b914688af3111ce19a427d80cd2f905920d815eac758dbcf416f","abstract_canon_sha256":"e7b9bf0198f11f8f4372d24b73d2e2b556271ff6d1c4917284c82f43c3a8cb38"},"schema_version":"1.0"},"canonical_sha256":"fe63ef469786eb027ae9e8e5e40bfd074c8b6af5812a60d283fce23687f0e2ec","source":{"kind":"arxiv","id":"1806.10230","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1806.10230","created_at":"2026-05-17T23:43:43Z"},{"alias_kind":"arxiv_version","alias_value":"1806.10230v4","created_at":"2026-05-17T23:43:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1806.10230","created_at":"2026-05-17T23:43:43Z"},{"alias_kind":"pith_short_12","alias_value":"7ZR66RUXQ3VQ","created_at":"2026-05-18T12:32:13Z"},{"alias_kind":"pith_short_16","alias_value":"7ZR66RUXQ3VQE6XJ","created_at":"2026-05-18T12:32:13Z"},{"alias_kind":"pith_short_8","alias_value":"7ZR66RUX","created_at":"2026-05-18T12:32:13Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:7ZR66RUXQ3VQE6XJ5DS6IC75A5","target":"record","payload":{"canonical_record":{"source":{"id":"1806.10230","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NE","submitted_at":"2018-06-26T22:14:36Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"5d87f51ac364b914688af3111ce19a427d80cd2f905920d815eac758dbcf416f","abstract_canon_sha256":"e7b9bf0198f11f8f4372d24b73d2e2b556271ff6d1c4917284c82f43c3a8cb38"},"schema_version":"1.0"},"canonical_sha256":"fe63ef469786eb027ae9e8e5e40bfd074c8b6af5812a60d283fce23687f0e2ec","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:43:43.203582Z","signature_b64":"MZZQacMmTtsYpS5d1S3yjC6Tt33neZbXvJfEaKoV+5ifomotHLmHNl664p25Q+KnvnJN7LJvLfdJB8tjq+58Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fe63ef469786eb027ae9e8e5e40bfd074c8b6af5812a60d283fce23687f0e2ec","last_reissued_at":"2026-05-17T23:43:43.202607Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:43:43.202607Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1806.10230","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:43:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zhvrzhONgMgoSJ8BlWyRElOpiNogKVu4geu5FvJPY6/Fm/NN6WJDxS/jEmW8A14JL8lnv1Ii23enaFS5zmJSAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T08:59:20.375168Z"},"content_sha256":"1bf96b3c1410fba8fde393f00fbcc688c2e62d45c456a423f04c6be8873c99b1","schema_version":"1.0","event_id":"sha256:1bf96b3c1410fba8fde393f00fbcc688c2e62d45c456a423f04c6be8873c99b1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:7ZR66RUXQ3VQE6XJ5DS6IC75A5","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Guided evolutionary strategies: Augmenting random search with surrogate gradients","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cs.NE","authors_text":"Dami Choi, George Tucker, Jascha Sohl-Dickstein, Luke Metz, Niru Maheswaranathan","submitted_at":"2018-06-26T22:14:36Z","abstract_excerpt":"Many applications in machine learning require optimizing a function whose true gradient is unknown, but where surrogate gradient information (directions that may be correlated with, but not necessarily identical to, the true gradient) is available instead. This arises when an approximate gradient is easier to compute than the full gradient (e.g. in meta-learning or unrolled optimization), or when a true gradient is intractable and is replaced with a surrogate (e.g. in certain reinforcement learning applications, or when using synthetic gradients). We propose Guided Evolutionary Strategies, a m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1806.10230","kind":"arxiv","version":4},"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":"2026-05-17T23:43:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rYfcBevinwlwGUdT+ynVDBGddlFAFlKzuytojCI8ZE+qVVTnTRErioT/G6UO2p41Co+Z0gB8Hj/4e6YbTDOZCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T08:59:20.375590Z"},"content_sha256":"11c4cb77681fb871795b0920dc83aba6439f991983c83dc402a9ea81c2bf4828","schema_version":"1.0","event_id":"sha256:11c4cb77681fb871795b0920dc83aba6439f991983c83dc402a9ea81c2bf4828"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7ZR66RUXQ3VQE6XJ5DS6IC75A5/bundle.json","state_url":"https://pith.science/pith/7ZR66RUXQ3VQE6XJ5DS6IC75A5/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7ZR66RUXQ3VQE6XJ5DS6IC75A5/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":"2026-05-27T08:59:20Z","links":{"resolver":"https://pith.science/pith/7ZR66RUXQ3VQE6XJ5DS6IC75A5","bundle":"https://pith.science/pith/7ZR66RUXQ3VQE6XJ5DS6IC75A5/bundle.json","state":"https://pith.science/pith/7ZR66RUXQ3VQE6XJ5DS6IC75A5/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7ZR66RUXQ3VQE6XJ5DS6IC75A5/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:7ZR66RUXQ3VQE6XJ5DS6IC75A5","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":"e7b9bf0198f11f8f4372d24b73d2e2b556271ff6d1c4917284c82f43c3a8cb38","cross_cats_sorted":["cs.LG","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NE","submitted_at":"2018-06-26T22:14:36Z","title_canon_sha256":"5d87f51ac364b914688af3111ce19a427d80cd2f905920d815eac758dbcf416f"},"schema_version":"1.0","source":{"id":"1806.10230","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1806.10230","created_at":"2026-05-17T23:43:43Z"},{"alias_kind":"arxiv_version","alias_value":"1806.10230v4","created_at":"2026-05-17T23:43:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1806.10230","created_at":"2026-05-17T23:43:43Z"},{"alias_kind":"pith_short_12","alias_value":"7ZR66RUXQ3VQ","created_at":"2026-05-18T12:32:13Z"},{"alias_kind":"pith_short_16","alias_value":"7ZR66RUXQ3VQE6XJ","created_at":"2026-05-18T12:32:13Z"},{"alias_kind":"pith_short_8","alias_value":"7ZR66RUX","created_at":"2026-05-18T12:32:13Z"}],"graph_snapshots":[{"event_id":"sha256:11c4cb77681fb871795b0920dc83aba6439f991983c83dc402a9ea81c2bf4828","target":"graph","created_at":"2026-05-17T23:43:43Z","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":"Many applications in machine learning require optimizing a function whose true gradient is unknown, but where surrogate gradient information (directions that may be correlated with, but not necessarily identical to, the true gradient) is available instead. This arises when an approximate gradient is easier to compute than the full gradient (e.g. in meta-learning or unrolled optimization), or when a true gradient is intractable and is replaced with a surrogate (e.g. in certain reinforcement learning applications, or when using synthetic gradients). We propose Guided Evolutionary Strategies, a m","authors_text":"Dami Choi, George Tucker, Jascha Sohl-Dickstein, Luke Metz, Niru Maheswaranathan","cross_cats":["cs.LG","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NE","submitted_at":"2018-06-26T22:14:36Z","title":"Guided evolutionary strategies: Augmenting random search with surrogate gradients"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1806.10230","kind":"arxiv","version":4},"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":"sha256:1bf96b3c1410fba8fde393f00fbcc688c2e62d45c456a423f04c6be8873c99b1","target":"record","created_at":"2026-05-17T23:43:43Z","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":"e7b9bf0198f11f8f4372d24b73d2e2b556271ff6d1c4917284c82f43c3a8cb38","cross_cats_sorted":["cs.LG","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NE","submitted_at":"2018-06-26T22:14:36Z","title_canon_sha256":"5d87f51ac364b914688af3111ce19a427d80cd2f905920d815eac758dbcf416f"},"schema_version":"1.0","source":{"id":"1806.10230","kind":"arxiv","version":4}},"canonical_sha256":"fe63ef469786eb027ae9e8e5e40bfd074c8b6af5812a60d283fce23687f0e2ec","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"fe63ef469786eb027ae9e8e5e40bfd074c8b6af5812a60d283fce23687f0e2ec","first_computed_at":"2026-05-17T23:43:43.202607Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:43:43.202607Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"MZZQacMmTtsYpS5d1S3yjC6Tt33neZbXvJfEaKoV+5ifomotHLmHNl664p25Q+KnvnJN7LJvLfdJB8tjq+58Dw==","signature_status":"signed_v1","signed_at":"2026-05-17T23:43:43.203582Z","signed_message":"canonical_sha256_bytes"},"source_id":"1806.10230","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1bf96b3c1410fba8fde393f00fbcc688c2e62d45c456a423f04c6be8873c99b1","sha256:11c4cb77681fb871795b0920dc83aba6439f991983c83dc402a9ea81c2bf4828"],"state_sha256":"627202bb870b98cb4193f7f01e64b5eb5e2fd657ad068aa2ceaaf3378a18a451"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ulF0psrFsZMX0f+pguyIYlMVX2GWW7oX0Xbx91McSyJtpD6ViV4/8EA+lzPqIh4qeiHPdlyuHzzyMk3Mw7mGBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-27T08:59:20.378325Z","bundle_sha256":"19e7018b8f32c2d5f8d4fdf8ee6d6a77392e0dd23f5451385d9de628d10300ba"}}