{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:O6G5VGSTXX7SVS4LKUZDA2OQBK","short_pith_number":"pith:O6G5VGST","schema_version":"1.0","canonical_sha256":"778dda9a53bdff2acb8b55323069d00a815da284f1672582c419c4e53a02dc9d","source":{"kind":"arxiv","id":"2103.03076","version":2},"attestation_state":"computed","paper":{"title":"Dynamic Efficient Adversarial Training Guided by Gradient Magnitude","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR"],"primary_cat":"cs.LG","authors_text":"Fu Wang, Wenjie Ruan, Yanbin Zheng, Yanghao Zhang","submitted_at":"2021-03-04T14:57:53Z","abstract_excerpt":"Adversarial training is an effective but time-consuming way to train robust deep neural networks that can withstand strong adversarial attacks. As a response to its inefficiency, we propose Dynamic Efficient Adversarial Training (DEAT), which gradually increases the adversarial iteration during training. We demonstrate that the gradient's magnitude correlates with the curvature of the trained model's loss landscape, allowing it to reflect the effect of adversarial training. Therefore, based on the magnitude of the gradient, we propose a general acceleration strategy, M+ acceleration, which ena"},"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":"2103.03076","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-03-04T14:57:53Z","cross_cats_sorted":["cs.CR"],"title_canon_sha256":"094586fb2b99296003c5079264ab21261fa25d44d96707bf404f95a218f7221a","abstract_canon_sha256":"c56882720344d6dddcf072afa6b4fa12fb223958933007367174f9bd24b65790"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:50:37.141416Z","signature_b64":"6cYHJz2NCwyMV3akbRntZFmvkJHG4IP0DG2HRXWS8Acr7gLeQdxrTfHICKspSBWtm0hnRikuCs3EhlYV0oT3AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"778dda9a53bdff2acb8b55323069d00a815da284f1672582c419c4e53a02dc9d","last_reissued_at":"2026-07-05T05:50:37.140933Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:50:37.140933Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Dynamic Efficient Adversarial Training Guided by Gradient Magnitude","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR"],"primary_cat":"cs.LG","authors_text":"Fu Wang, Wenjie Ruan, Yanbin Zheng, Yanghao Zhang","submitted_at":"2021-03-04T14:57:53Z","abstract_excerpt":"Adversarial training is an effective but time-consuming way to train robust deep neural networks that can withstand strong adversarial attacks. As a response to its inefficiency, we propose Dynamic Efficient Adversarial Training (DEAT), which gradually increases the adversarial iteration during training. We demonstrate that the gradient's magnitude correlates with the curvature of the trained model's loss landscape, allowing it to reflect the effect of adversarial training. Therefore, based on the magnitude of the gradient, we propose a general acceleration strategy, M+ acceleration, which ena"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2103.03076","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/2103.03076/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":"2103.03076","created_at":"2026-07-05T05:50:37.140992+00:00"},{"alias_kind":"arxiv_version","alias_value":"2103.03076v2","created_at":"2026-07-05T05:50:37.140992+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2103.03076","created_at":"2026-07-05T05:50:37.140992+00:00"},{"alias_kind":"pith_short_12","alias_value":"O6G5VGSTXX7S","created_at":"2026-07-05T05:50:37.140992+00:00"},{"alias_kind":"pith_short_16","alias_value":"O6G5VGSTXX7SVS4L","created_at":"2026-07-05T05:50:37.140992+00:00"},{"alias_kind":"pith_short_8","alias_value":"O6G5VGST","created_at":"2026-07-05T05:50:37.140992+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/O6G5VGSTXX7SVS4LKUZDA2OQBK","json":"https://pith.science/pith/O6G5VGSTXX7SVS4LKUZDA2OQBK.json","graph_json":"https://pith.science/api/pith-number/O6G5VGSTXX7SVS4LKUZDA2OQBK/graph.json","events_json":"https://pith.science/api/pith-number/O6G5VGSTXX7SVS4LKUZDA2OQBK/events.json","paper":"https://pith.science/paper/O6G5VGST"},"agent_actions":{"view_html":"https://pith.science/pith/O6G5VGSTXX7SVS4LKUZDA2OQBK","download_json":"https://pith.science/pith/O6G5VGSTXX7SVS4LKUZDA2OQBK.json","view_paper":"https://pith.science/paper/O6G5VGST","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2103.03076&json=true","fetch_graph":"https://pith.science/api/pith-number/O6G5VGSTXX7SVS4LKUZDA2OQBK/graph.json","fetch_events":"https://pith.science/api/pith-number/O6G5VGSTXX7SVS4LKUZDA2OQBK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/O6G5VGSTXX7SVS4LKUZDA2OQBK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/O6G5VGSTXX7SVS4LKUZDA2OQBK/action/storage_attestation","attest_author":"https://pith.science/pith/O6G5VGSTXX7SVS4LKUZDA2OQBK/action/author_attestation","sign_citation":"https://pith.science/pith/O6G5VGSTXX7SVS4LKUZDA2OQBK/action/citation_signature","submit_replication":"https://pith.science/pith/O6G5VGSTXX7SVS4LKUZDA2OQBK/action/replication_record"}},"created_at":"2026-07-05T05:50:37.140992+00:00","updated_at":"2026-07-05T05:50:37.140992+00:00"}