{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:7ZHYENLLAORV5CMBTEYHUCT7JP","short_pith_number":"pith:7ZHYENLL","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"},"canonical_sha256":"fe4f82356b03a35e898199307a0a7f4bf960794febaf22270ec2e91b5b6d1185","source":{"kind":"arxiv","id":"2203.09962","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2203.09962","created_at":"2026-07-05T05:59:20Z"},{"alias_kind":"arxiv_version","alias_value":"2203.09962v2","created_at":"2026-07-05T05:59:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.09962","created_at":"2026-07-05T05:59:20Z"},{"alias_kind":"pith_short_12","alias_value":"7ZHYENLLAORV","created_at":"2026-07-05T05:59:20Z"},{"alias_kind":"pith_short_16","alias_value":"7ZHYENLLAORV5CMB","created_at":"2026-07-05T05:59:20Z"},{"alias_kind":"pith_short_8","alias_value":"7ZHYENLL","created_at":"2026-07-05T05:59:20Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:7ZHYENLLAORV5CMBTEYHUCT7JP","target":"record","payload":{"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"},"canonical_sha256":"fe4f82356b03a35e898199307a0a7f4bf960794febaf22270ec2e91b5b6d1185","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"},"source_kind":"arxiv","source_id":"2203.09962","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T05:59:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7LdfmyT7SJ9+bRjfXQrMZpy85FThrjiasNTOgi9h8OQ3LxGnvI7Hyq8ZCT3iMhYJNnzyWGQsa+CT77m3NOVRDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T06:51:21.652311Z"},"content_sha256":"944eee7d4f06cbcb1c09549bc17d9e942d48c317edae6a53ea13afcc38364aca","schema_version":"1.0","event_id":"sha256:944eee7d4f06cbcb1c09549bc17d9e942d48c317edae6a53ea13afcc38364aca"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:7ZHYENLLAORV5CMBTEYHUCT7JP","target":"graph","payload":{"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T05:59:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lqNnTYDjOpIIgIeZGxf22PIilLbZUGnaulzx+GvempgVHyt03u0gkw/wBj6ydTz/Sw/FUjHFiQHlR/YamUhIDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T06:51:21.652927Z"},"content_sha256":"7caa20b36f2693779feb7a27004009981db82e3b0b190bce57b2a6d8e8046d33","schema_version":"1.0","event_id":"sha256:7caa20b36f2693779feb7a27004009981db82e3b0b190bce57b2a6d8e8046d33"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7ZHYENLLAORV5CMBTEYHUCT7JP/bundle.json","state_url":"https://pith.science/pith/7ZHYENLLAORV5CMBTEYHUCT7JP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7ZHYENLLAORV5CMBTEYHUCT7JP/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":"2026-07-07T06:51:21Z","links":{"resolver":"https://pith.science/pith/7ZHYENLLAORV5CMBTEYHUCT7JP","bundle":"https://pith.science/pith/7ZHYENLLAORV5CMBTEYHUCT7JP/bundle.json","state":"https://pith.science/pith/7ZHYENLLAORV5CMBTEYHUCT7JP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7ZHYENLLAORV5CMBTEYHUCT7JP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:7ZHYENLLAORV5CMBTEYHUCT7JP","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":"4e41f96d68ab6de9c482bd4e084a94615807b38984185d4a6903ea37ad5ddcd6","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-03-18T13:57:17Z","title_canon_sha256":"10eba1685a05a74b22c431abebf679b7ed4e92029db390c11dd8455dbd820467"},"schema_version":"1.0","source":{"id":"2203.09962","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2203.09962","created_at":"2026-07-05T05:59:20Z"},{"alias_kind":"arxiv_version","alias_value":"2203.09962v2","created_at":"2026-07-05T05:59:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.09962","created_at":"2026-07-05T05:59:20Z"},{"alias_kind":"pith_short_12","alias_value":"7ZHYENLLAORV","created_at":"2026-07-05T05:59:20Z"},{"alias_kind":"pith_short_16","alias_value":"7ZHYENLLAORV5CMB","created_at":"2026-07-05T05:59:20Z"},{"alias_kind":"pith_short_8","alias_value":"7ZHYENLL","created_at":"2026-07-05T05:59:20Z"}],"graph_snapshots":[{"event_id":"sha256:7caa20b36f2693779feb7a27004009981db82e3b0b190bce57b2a6d8e8046d33","target":"graph","created_at":"2026-07-05T05:59:20Z","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/2203.09962/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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","authors_text":"Hao Zhang, Xiuyuan Hu, Yang Zhao","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-03-18T13:57:17Z","title":"Randomized Sharpness-Aware Training for Boosting Computational Efficiency in Deep Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.09962","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":"sha256:944eee7d4f06cbcb1c09549bc17d9e942d48c317edae6a53ea13afcc38364aca","target":"record","created_at":"2026-07-05T05:59:20Z","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":"4e41f96d68ab6de9c482bd4e084a94615807b38984185d4a6903ea37ad5ddcd6","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-03-18T13:57:17Z","title_canon_sha256":"10eba1685a05a74b22c431abebf679b7ed4e92029db390c11dd8455dbd820467"},"schema_version":"1.0","source":{"id":"2203.09962","kind":"arxiv","version":2}},"canonical_sha256":"fe4f82356b03a35e898199307a0a7f4bf960794febaf22270ec2e91b5b6d1185","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"fe4f82356b03a35e898199307a0a7f4bf960794febaf22270ec2e91b5b6d1185","first_computed_at":"2026-07-05T05:59:20.240387Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:59:20.240387Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"S+AR3SBbnIoJTzf2jRsSFaQokU+BfrWk2C/KFY6HyxzxUoJjka4GHiRm9eHdcifHpmneg0eJVz9dUEJ7QPmvBw==","signature_status":"signed_v1","signed_at":"2026-07-05T05:59:20.240843Z","signed_message":"canonical_sha256_bytes"},"source_id":"2203.09962","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:944eee7d4f06cbcb1c09549bc17d9e942d48c317edae6a53ea13afcc38364aca","sha256:7caa20b36f2693779feb7a27004009981db82e3b0b190bce57b2a6d8e8046d33"],"state_sha256":"d3f51bcf9f7a2a6d9c20d392e517c8bbca46e0974190c898ccbc4580f90cd38d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TuP+pp786UPft80XaHOaXH0EBK7U2iONBFAKZoHEpG3aJ8nb04qAk0WVSC7f1ZUoqmv+/51Wizp0xbtE33qTCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T06:51:21.655675Z","bundle_sha256":"0a2e660875d10e4377e7b0fdaed0bfd96764a7c7cf86b34fa2e67bd1adec77d4"}}