{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:7ZHYENLLAORV5CMBTEYHUCT7JP","short_pith_number":"pith:7ZHYENLL","schema_version":"1.0","canonical_sha256":"fe4f82356b03a35e898199307a0a7f4bf960794febaf22270ec2e91b5b6d1185","source":{"kind":"arxiv","id":"2203.09962","version":2},"attestation_state":"computed","paper":{"title":"Randomized Sharpness-Aware Training for Boosting Computational Efficiency in Deep Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Hao Zhang, Xiuyuan Hu, Yang Zhao","submitted_at":"2022-03-18T13:57:17Z","abstract_excerpt":"By driving models to converge to flat minima, sharpness-aware learning algorithms (such as SAM) have shown the power to achieve state-of-the-art performances. However, these algorithms will generally incur one extra forward-backward propagation at each training iteration, which largely burdens the computation especially for scalable models. To this end, we propose a simple yet efficient training scheme, called Randomized Sharpness-Aware Training (RST). Optimizers in RST would perform a Bernoulli trial at each iteration to choose randomly from base algorithms (SGD) and sharpness-aware algorithm"},"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":"2203.09962","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-03-18T13:57:17Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"10eba1685a05a74b22c431abebf679b7ed4e92029db390c11dd8455dbd820467","abstract_canon_sha256":"4e41f96d68ab6de9c482bd4e084a94615807b38984185d4a6903ea37ad5ddcd6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:59:20.240843Z","signature_b64":"S+AR3SBbnIoJTzf2jRsSFaQokU+BfrWk2C/KFY6HyxzxUoJjka4GHiRm9eHdcifHpmneg0eJVz9dUEJ7QPmvBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fe4f82356b03a35e898199307a0a7f4bf960794febaf22270ec2e91b5b6d1185","last_reissued_at":"2026-07-05T05:59:20.240387Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:59:20.240387Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Randomized Sharpness-Aware Training for Boosting Computational Efficiency in Deep Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Hao Zhang, Xiuyuan Hu, Yang Zhao","submitted_at":"2022-03-18T13:57:17Z","abstract_excerpt":"By driving models to converge to flat minima, sharpness-aware learning algorithms (such as SAM) have shown the power to achieve state-of-the-art performances. However, these algorithms will generally incur one extra forward-backward propagation at each training iteration, which largely burdens the computation especially for scalable models. To this end, we propose a simple yet efficient training scheme, called Randomized Sharpness-Aware Training (RST). Optimizers in RST would perform a Bernoulli trial at each iteration to choose randomly from base algorithms (SGD) and sharpness-aware algorithm"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.09962","kind":"arxiv","version":2},"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/2203.09962/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":"2203.09962","created_at":"2026-07-05T05:59:20.240446+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.09962v2","created_at":"2026-07-05T05:59:20.240446+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.09962","created_at":"2026-07-05T05:59:20.240446+00:00"},{"alias_kind":"pith_short_12","alias_value":"7ZHYENLLAORV","created_at":"2026-07-05T05:59:20.240446+00:00"},{"alias_kind":"pith_short_16","alias_value":"7ZHYENLLAORV5CMB","created_at":"2026-07-05T05:59:20.240446+00:00"},{"alias_kind":"pith_short_8","alias_value":"7ZHYENLL","created_at":"2026-07-05T05:59:20.240446+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.11064","citing_title":"A Faster Path to Continual Learning","ref_index":63,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7ZHYENLLAORV5CMBTEYHUCT7JP","json":"https://pith.science/pith/7ZHYENLLAORV5CMBTEYHUCT7JP.json","graph_json":"https://pith.science/api/pith-number/7ZHYENLLAORV5CMBTEYHUCT7JP/graph.json","events_json":"https://pith.science/api/pith-number/7ZHYENLLAORV5CMBTEYHUCT7JP/events.json","paper":"https://pith.science/paper/7ZHYENLL"},"agent_actions":{"view_html":"https://pith.science/pith/7ZHYENLLAORV5CMBTEYHUCT7JP","download_json":"https://pith.science/pith/7ZHYENLLAORV5CMBTEYHUCT7JP.json","view_paper":"https://pith.science/paper/7ZHYENLL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.09962&json=true","fetch_graph":"https://pith.science/api/pith-number/7ZHYENLLAORV5CMBTEYHUCT7JP/graph.json","fetch_events":"https://pith.science/api/pith-number/7ZHYENLLAORV5CMBTEYHUCT7JP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7ZHYENLLAORV5CMBTEYHUCT7JP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7ZHYENLLAORV5CMBTEYHUCT7JP/action/storage_attestation","attest_author":"https://pith.science/pith/7ZHYENLLAORV5CMBTEYHUCT7JP/action/author_attestation","sign_citation":"https://pith.science/pith/7ZHYENLLAORV5CMBTEYHUCT7JP/action/citation_signature","submit_replication":"https://pith.science/pith/7ZHYENLLAORV5CMBTEYHUCT7JP/action/replication_record"}},"created_at":"2026-07-05T05:59:20.240446+00:00","updated_at":"2026-07-05T05:59:20.240446+00:00"}