{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/scaling-adversarial-training-to-large","title":"Scaling Adversarial Training to Large Perturbation Bounds","arxiv_id":"2210.09852","date":"2022-10-18","proceeding":null,"authors":["Sravanti Addepalli","Samyak Jain","Gaurang Sriramanan","R. Venkatesh Babu"],"abstract":"The vulnerability of Deep Neural Networks to Adversarial Attacks has fuelled research towards building robust models. While most Adversarial Training algorithms aim at defending attacks constrained within low magnitude Lp norm bounds, real-world adversaries are not limited by such constraints. In this work, we aim to achieve adversarial robustness within larger bounds, against perturbations that may be perceptible, but do not change human (or Oracle) prediction. The presence of images that flip Oracle predictions and those that do not makes this a challenging setting for adversarial robustness. We discuss the ideal goals of an adversarial defense algorithm beyond perceptual limits, and further highlight the shortcomings of naively extending existing training algorithms to higher perturbation bounds. In order to overcome these shortcomings, we propose a novel defense, Oracle-Aligned Adversarial Training (OA-AT), to align the predictions of the network with that of an Oracle during adversarial training. The proposed approach achieves state-of-the-art performance at large epsilon bounds (such as an L-inf bound of 16/255 on CIFAR-10) while outperforming existing defenses (AWP, TRADES, PGD-AT) at standard bounds (8/255) as well.","url_abs":"https://arxiv.org/abs/2210.09852v1","url_pdf":"https://arxiv.org/pdf/2210.09852v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"scaling-adversarial-training-to-large","repo_url":"https://github.com/val-iisc/oaat","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"adversarial-defense","task_name":"Adversarial Defense"},{"task_slug":"adversarial-robustness","task_name":"Adversarial Robustness"}],"methods":[{"method_slug":"align","method_name":"ALIGN"},{"method_slug":"flip","method_name":"FLIP"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2210.09852","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2210.09852"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/val-iisc/oaat","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"deterministic:regex_extraction","url":"https://github.com/val-iisc/OAAT","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":3,"ran_draft_wrong":1,"unverified":4},"by_repo_kind":{"official":{"samples":8,"ran":4,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"7fad0aef219ec2a6","entry":"create_window","repo":"val-iisc/OAAT","repo_kind":"official","path":"perceptual_advex/distances.py","file_url":"https://github.com/val-iisc/OAAT/blob/HEAD/perceptual_advex/distances.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"7fad0aef219ec2a6"}},{"code_sha256_prefix":"4170809219439d13","entry":"diff_in_weights","repo":"val-iisc/OAAT","repo_kind":"official","path":"utils_awp.py","file_url":"https://github.com/val-iisc/OAAT/blob/HEAD/utils_awp.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"4170809219439d13"}},{"code_sha256_prefix":"a945dca82afbbec0","entry":"gaussian","repo":"val-iisc/OAAT","repo_kind":"official","path":"perceptual_advex/distances.py","file_url":"https://github.com/val-iisc/OAAT/blob/HEAD/perceptual_advex/distances.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a945dca82afbbec0"}},{"code_sha256_prefix":"096fecc8a890169a","entry":"normalize_flatten_features","repo":"val-iisc/OAAT","repo_kind":"official","path":"perceptual_advex/distances.py","file_url":"https://github.com/val-iisc/OAAT/blob/HEAD/perceptual_advex/distances.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"096fecc8a890169a"}},{"code_sha256_prefix":"6720f86e06ca335e","entry":"PreActResNet18","repo":"val-iisc/OAAT","repo_kind":"official","path":"models/preactresnet.py","file_url":"https://github.com/val-iisc/OAAT/blob/HEAD/models/preactresnet.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"6720f86e06ca335e"}},{"code_sha256_prefix":"5b31923638042434","entry":"ResNet18","repo":"val-iisc/OAAT","repo_kind":"official","path":"models/resnet.py","file_url":"https://github.com/val-iisc/OAAT/blob/HEAD/models/resnet.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"5b31923638042434"}},{"code_sha256_prefix":"f847e64ed88ffc03","entry":"WideResNet34","repo":"val-iisc/OAAT","repo_kind":"official","path":"models/wideresnet.py","file_url":"https://github.com/val-iisc/OAAT/blob/HEAD/models/wideresnet.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"f847e64ed88ffc03"}},{"code_sha256_prefix":"ca563345f2d2b46e","entry":"use_default","repo":"val-iisc/OAAT","repo_kind":"official","path":"defaults.py","file_url":"https://github.com/val-iisc/OAAT/blob/HEAD/defaults.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"ca563345f2d2b46e"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}