{"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/exploring-and-exploiting-decision-boundary","title":"Exploring and Exploiting Decision Boundary Dynamics for Adversarial Robustness","arxiv_id":"2302.03015","date":"2023-02-06","proceeding":null,"authors":["Yuancheng Xu","Yanchao Sun","Micah Goldblum","Tom Goldstein","Furong Huang"],"abstract":"The robustness of a deep classifier can be characterized by its margins: the decision boundary's distances to natural data points. However, it is unclear whether existing robust training methods effectively increase the margin for each vulnerable point during training. To understand this, we propose a continuous-time framework for quantifying the relative speed of the decision boundary with respect to each individual point. Through visualizing the moving speed of the decision boundary under Adversarial Training, one of the most effective robust training algorithms, a surprising moving-behavior is revealed: the decision boundary moves away from some vulnerable points but simultaneously moves closer to others, decreasing their margins. To alleviate these conflicting dynamics of the decision boundary, we propose Dynamics-aware Robust Training (DyART), which encourages the decision boundary to engage in movement that prioritizes increasing smaller margins. In contrast to prior works, DyART directly operates on the margins rather than their indirect approximations, allowing for more targeted and effective robustness improvement. Experiments on the CIFAR-10 and Tiny-ImageNet datasets verify that DyART alleviates the conflicting dynamics of the decision boundary and obtains improved robustness under various perturbation sizes compared to the state-of-the-art defenses. Our code is available at https://github.com/Yuancheng-Xu/Dynamics-Aware-Robust-Training.","url_abs":"https://arxiv.org/abs/2302.03015v2","url_pdf":"https://arxiv.org/pdf/2302.03015v2.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":"exploring-and-exploiting-decision-boundary","repo_url":"https://github.com/yuancheng-xu/dynamics-aware-robust-training","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"exploring-and-exploiting-decision-boundary","repo_url":"https://github.com/umd-huang-lab/vlm-poisoning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"adversarial-robustness","task_name":"Adversarial Robustness"}],"methods":[{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2302.03015","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2302.03015"}},"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/umd-huang-lab/vlm-poisoning","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/yuancheng-xu/dynamics-aware-robust-training","reach":null}],"summary":{"ran":2,"ran_honours":1,"ran_violates":1,"unverified":8},"by_repo_kind":{"official":{"samples":12,"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":"b5c1ada412a147f0","entry":"Attack","repo":"yuancheng-xu/dynamics-aware-robust-training","repo_kind":"official","path":"core/DyART/DyART_Linf.py","file_url":"https://github.com/yuancheng-xu/dynamics-aware-robust-training/blob/HEAD/core/DyART/DyART_Linf.py","link_basis":"first_harvest_node","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":"b5c1ada412a147f0"}},{"code_sha256_prefix":"54c90b4b9c297c04","entry":"FABAttack_scalar","repo":"yuancheng-xu/dynamics-aware-robust-training","repo_kind":"official","path":"core/DyART/DyART_Linf.py","file_url":"https://github.com/yuancheng-xu/dynamics-aware-robust-training/blob/HEAD/core/DyART/DyART_Linf.py","link_basis":"first_harvest_node","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":"54c90b4b9c297c04"}},{"code_sha256_prefix":"4e78e41a8926dc30","entry":"replicate_input","repo":"yuancheng-xu/dynamics-aware-robust-training","repo_kind":"official","path":"core/DyART/DyART_Linf.py","file_url":"https://github.com/yuancheng-xu/dynamics-aware-robust-training/blob/HEAD/core/DyART/DyART_Linf.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"4e78e41a8926dc30"}},{"code_sha256_prefix":"50714c3786441e94","entry":"smooth_max","repo":"yuancheng-xu/dynamics-aware-robust-training","repo_kind":"official","path":"core/DyART/DyART_Linf.py","file_url":"https://github.com/yuancheng-xu/dynamics-aware-robust-training/blob/HEAD/core/DyART/DyART_Linf.py","link_basis":"first_harvest_node","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"50714c3786441e94"}},{"code_sha256_prefix":"23fb5907f52402c5","entry":"DyART_loss_Linf","repo":"yuancheng-xu/dynamics-aware-robust-training","repo_kind":"official","path":"core/DyART/DyART_Linf.py","file_url":"https://github.com/yuancheng-xu/dynamics-aware-robust-training/blob/HEAD/core/DyART/DyART_Linf.py","link_basis":"first_harvest_node","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":"23fb5907f52402c5"}},{"code_sha256_prefix":"8ed3a2aa6ae2a7ba","entry":"LabelMixin","repo":"yuancheng-xu/dynamics-aware-robust-training","repo_kind":"official","path":"core/DyART/DyART_Linf.py","file_url":"https://github.com/yuancheng-xu/dynamics-aware-robust-training/blob/HEAD/core/DyART/DyART_Linf.py","link_basis":"first_harvest_node","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":"8ed3a2aa6ae2a7ba"}},{"code_sha256_prefix":"1dbfdf8db17c1adf","entry":"find_bdr_FAB","repo":"yuancheng-xu/dynamics-aware-robust-training","repo_kind":"official","path":"core/DyART/DyART_Linf.py","file_url":"https://github.com/yuancheng-xu/dynamics-aware-robust-training/blob/HEAD/core/DyART/DyART_Linf.py","link_basis":"first_harvest_node","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":"1dbfdf8db17c1adf"}},{"code_sha256_prefix":"e406ff37047a4a9f","entry":"find_bdr_Linf","repo":"yuancheng-xu/dynamics-aware-robust-training","repo_kind":"official","path":"core/DyART/DyART_Linf.py","file_url":"https://github.com/yuancheng-xu/dynamics-aware-robust-training/blob/HEAD/core/DyART/DyART_Linf.py","link_basis":"first_harvest_node","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":"e406ff37047a4a9f"}},{"code_sha256_prefix":"15f8f0967b36aa06","entry":"func_soft_margin","repo":"yuancheng-xu/dynamics-aware-robust-training","repo_kind":"official","path":"core/DyART/DyART_Linf.py","file_url":"https://github.com/yuancheng-xu/dynamics-aware-robust-training/blob/HEAD/core/DyART/DyART_Linf.py","link_basis":"first_harvest_node","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":"15f8f0967b36aa06"}},{"code_sha256_prefix":"a1e687f30713373e","entry":"get_radius_weights","repo":"yuancheng-xu/dynamics-aware-robust-training","repo_kind":"official","path":"core/DyART/DyART_Linf.py","file_url":"https://github.com/yuancheng-xu/dynamics-aware-robust-training/blob/HEAD/core/DyART/DyART_Linf.py","link_basis":"first_harvest_node","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":"a1e687f30713373e"}},{"code_sha256_prefix":"9f6fad065902cd84","entry":"h_prime","repo":"yuancheng-xu/dynamics-aware-robust-training","repo_kind":"official","path":"core/DyART/DyART_Linf.py","file_url":"https://github.com/yuancheng-xu/dynamics-aware-robust-training/blob/HEAD/core/DyART/DyART_Linf.py","link_basis":"first_harvest_node","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":"9f6fad065902cd84"}},{"code_sha256_prefix":"9522fba05d4568aa","entry":"zero_gradients","repo":"yuancheng-xu/dynamics-aware-robust-training","repo_kind":"official","path":"core/DyART/DyART_Linf.py","file_url":"https://github.com/yuancheng-xu/dynamics-aware-robust-training/blob/HEAD/core/DyART/DyART_Linf.py","link_basis":"first_harvest_node","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":"9522fba05d4568aa"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}