{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:KJZHVAJCPF74OOYI4QF3CGD7UY","short_pith_number":"pith:KJZHVAJC","schema_version":"1.0","canonical_sha256":"52727a8122797fc73b08e40bb1187fa625f0dee9cb7405db07d3b7b1f1ba37da","source":{"kind":"arxiv","id":"2310.02025","version":4},"attestation_state":"computed","paper":{"title":"DeepZero: Scaling up Zeroth-Order Optimization for Deep Model Training","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Aochuan Chen, Bhavya Kailkhura, James Diffenderfer, Jiancheng Liu, Jinghan Jia, Konstantinos Parasyris, Sijia Liu, Yihua Zhang, Yimeng Zhang, Zheng Zhang","submitted_at":"2023-10-03T13:05:36Z","abstract_excerpt":"Zeroth-order (ZO) optimization has become a popular technique for solving machine learning (ML) problems when first-order (FO) information is difficult or impossible to obtain. However, the scalability of ZO optimization remains an open problem: Its use has primarily been limited to relatively small-scale ML problems, such as sample-wise adversarial attack generation. To our best knowledge, no prior work has demonstrated the effectiveness of ZO optimization in training deep neural networks (DNNs) without a significant decrease in performance. To overcome this roadblock, we develop DeepZero, a "},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2310.02025","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-10-03T13:05:36Z","cross_cats_sorted":[],"title_canon_sha256":"f3b1c51344ccca8bb827d812cbef87d52340fb9f7166eb30c59686d9a788f8ff","abstract_canon_sha256":"3056635826488f8444c8c5d470764f4001439c09dc5d4f9e83d8bc8dcd552a98"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:56:17.331840Z","signature_b64":"VZiGaNi0hDcGQ7wzoXviyF0FQgZkna2aX0kNpdsssPrn3Fab0bZo0sC7aXLmlaTRJ1GTk9dpgeDkWXFBNXtXAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"52727a8122797fc73b08e40bb1187fa625f0dee9cb7405db07d3b7b1f1ba37da","last_reissued_at":"2026-07-05T07:56:17.331112Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:56:17.331112Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DeepZero: Scaling up Zeroth-Order Optimization for Deep Model Training","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Aochuan Chen, Bhavya Kailkhura, James Diffenderfer, Jiancheng Liu, Jinghan Jia, Konstantinos Parasyris, Sijia Liu, Yihua Zhang, Yimeng Zhang, Zheng Zhang","submitted_at":"2023-10-03T13:05:36Z","abstract_excerpt":"Zeroth-order (ZO) optimization has become a popular technique for solving machine learning (ML) problems when first-order (FO) information is difficult or impossible to obtain. However, the scalability of ZO optimization remains an open problem: Its use has primarily been limited to relatively small-scale ML problems, such as sample-wise adversarial attack generation. To our best knowledge, no prior work has demonstrated the effectiveness of ZO optimization in training deep neural networks (DNNs) without a significant decrease in performance. To overcome this roadblock, we develop DeepZero, a "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.02025","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2310.02025/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2310.02025","created_at":"2026-07-05T07:56:17.331283+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.02025v4","created_at":"2026-07-05T07:56:17.331283+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.02025","created_at":"2026-07-05T07:56:17.331283+00:00"},{"alias_kind":"pith_short_12","alias_value":"KJZHVAJCPF74","created_at":"2026-07-05T07:56:17.331283+00:00"},{"alias_kind":"pith_short_16","alias_value":"KJZHVAJCPF74OOYI","created_at":"2026-07-05T07:56:17.331283+00:00"},{"alias_kind":"pith_short_8","alias_value":"KJZHVAJC","created_at":"2026-07-05T07:56:17.331283+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2601.17261","citing_title":"AGZO: Activation-Guided Zeroth-Order Optimization for LLM Fine-Tuning","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2509.15113","citing_title":"Low-rank surrogate modeling and stochastic zero-order optimization for training of neural networks with black-box layers","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00650","citing_title":"AdaMeZO: Adam-style Zeroth-Order Optimizer for LLM Fine-tuning Without Maintaining the Moments","ref_index":68,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KJZHVAJCPF74OOYI4QF3CGD7UY","json":"https://pith.science/pith/KJZHVAJCPF74OOYI4QF3CGD7UY.json","graph_json":"https://pith.science/api/pith-number/KJZHVAJCPF74OOYI4QF3CGD7UY/graph.json","events_json":"https://pith.science/api/pith-number/KJZHVAJCPF74OOYI4QF3CGD7UY/events.json","paper":"https://pith.science/paper/KJZHVAJC"},"agent_actions":{"view_html":"https://pith.science/pith/KJZHVAJCPF74OOYI4QF3CGD7UY","download_json":"https://pith.science/pith/KJZHVAJCPF74OOYI4QF3CGD7UY.json","view_paper":"https://pith.science/paper/KJZHVAJC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.02025&json=true","fetch_graph":"https://pith.science/api/pith-number/KJZHVAJCPF74OOYI4QF3CGD7UY/graph.json","fetch_events":"https://pith.science/api/pith-number/KJZHVAJCPF74OOYI4QF3CGD7UY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KJZHVAJCPF74OOYI4QF3CGD7UY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KJZHVAJCPF74OOYI4QF3CGD7UY/action/storage_attestation","attest_author":"https://pith.science/pith/KJZHVAJCPF74OOYI4QF3CGD7UY/action/author_attestation","sign_citation":"https://pith.science/pith/KJZHVAJCPF74OOYI4QF3CGD7UY/action/citation_signature","submit_replication":"https://pith.science/pith/KJZHVAJCPF74OOYI4QF3CGD7UY/action/replication_record"}},"created_at":"2026-07-05T07:56:17.331283+00:00","updated_at":"2026-07-05T07:56:17.331283+00:00"}