{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:CU2BVYF4KBKYIVCHIUB4SG6DLA","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":"4de2d1d5dda5997ed18ea24d69e5edc83f8148ab0d6af058b7a510091b746ab2","cross_cats_sorted":["cs.AI","cs.NE","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-06-02T06:03:18Z","title_canon_sha256":"d649198895a84d0e641524994a17423b081095e2c629248634ad7da533d71112"},"schema_version":"1.0","source":{"id":"2106.00958","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2106.00958","created_at":"2026-07-05T02:47:08Z"},{"alias_kind":"arxiv_version","alias_value":"2106.00958v2","created_at":"2026-07-05T02:47:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.00958","created_at":"2026-07-05T02:47:08Z"},{"alias_kind":"pith_short_12","alias_value":"CU2BVYF4KBKY","created_at":"2026-07-05T02:47:08Z"},{"alias_kind":"pith_short_16","alias_value":"CU2BVYF4KBKYIVCH","created_at":"2026-07-05T02:47:08Z"},{"alias_kind":"pith_short_8","alias_value":"CU2BVYF4","created_at":"2026-07-05T02:47:08Z"}],"graph_snapshots":[{"event_id":"sha256:3cdf81389f1133a4c95186c89c70600d5130719a676f84175d37390d04f5249c","target":"graph","created_at":"2026-07-05T02:47:08Z","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"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2106.00958/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"A core issue with learning to optimize neural networks has been the lack of generalization to real world problems. To address this, we describe a system designed from a generalization-first perspective, learning to update optimizer hyperparameters instead of model parameters directly using novel features, actions, and a reward function. This system outperforms Adam at all neural network tasks including on modalities not seen during training. We achieve 2x speedups on ImageNet, and a 2.5x speedup on a language modeling task using over 5 orders of magnitude more compute than the training tasks.","authors_text":"Clemens Winter, Diogo Almeida, Jie Tang, Wojciech Zaremba","cross_cats":["cs.AI","cs.NE","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-06-02T06:03:18Z","title":"A Generalizable Approach to Learning Optimizers"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.00958","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:deb8e939e39701acc064ee2cfde8be96dd50d2a4205293e02ab0333ed9bd7c90","target":"record","created_at":"2026-07-05T02:47:08Z","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":"4de2d1d5dda5997ed18ea24d69e5edc83f8148ab0d6af058b7a510091b746ab2","cross_cats_sorted":["cs.AI","cs.NE","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-06-02T06:03:18Z","title_canon_sha256":"d649198895a84d0e641524994a17423b081095e2c629248634ad7da533d71112"},"schema_version":"1.0","source":{"id":"2106.00958","kind":"arxiv","version":2}},"canonical_sha256":"15341ae0bc50558454474503c91bc35808017a293aff33e23cd5fbd1c219b090","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"15341ae0bc50558454474503c91bc35808017a293aff33e23cd5fbd1c219b090","first_computed_at":"2026-07-05T02:47:08.474312Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T02:47:08.474312Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ukfPkFuYdfgZbLo6DCuIuVK5yEQurioyjcs4GpkL8xlUjeajxpBA4yWObPxHA7OJmpgWOvYbeAQVxCVeXxzADA==","signature_status":"signed_v1","signed_at":"2026-07-05T02:47:08.474744Z","signed_message":"canonical_sha256_bytes"},"source_id":"2106.00958","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:deb8e939e39701acc064ee2cfde8be96dd50d2a4205293e02ab0333ed9bd7c90","sha256:3cdf81389f1133a4c95186c89c70600d5130719a676f84175d37390d04f5249c"],"state_sha256":"d0dc9a3ee017c05de221b2f6c9c03bffee30bd63f5f3f088f2b65334933aacde"}