{"about":{"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.","site":"https://codewithpapers.app","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","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/paper/a-new-family-of-neural-networks-provably","title":"A New Family of Neural Networks Provably Resistant to Adversarial Attacks","arxiv_id":"1902.01208","date":"2019-02-01","proceeding":null,"authors":["Rakshit Agrawal","Luca de Alfaro","David Helmbold"],"abstract":"Adversarial attacks add perturbations to the input features with the intent\nof changing the classification produced by a machine learning system. Small\nperturbations can yield adversarial examples which are misclassified despite\nbeing virtually indistinguishable from the unperturbed input. Classifiers\ntrained with standard neural network techniques are highly susceptible to\nadversarial examples, allowing an adversary to create misclassifications of\ntheir choice.\n  We introduce a new type of network unit, called MWD (max of weighed distance)\nunits that have a built-in resistant to adversarial attacks. These units are\nhighly non-linear, and we develop the techniques needed to effectively train\nthem. We show that simple interval techniques for propagating perturbation\neffects through the network enables the efficient computation of robustness\n(i.e., accuracy guarantees) for MWD networks under any perturbations, including\nadversarial attacks.\n  MWD networks are significantly more robust to input perturbations than ReLU\nnetworks. On permutation invariant MNIST, when test examples can be perturbed\nby 20% of the input range, MWD networks provably retain accuracy above 83%,\nwhile the accuracy of ReLU networks drops below 5%. The provable accuracy of\nMWD networks is superior even to the observed accuracy of ReLU networks trained\nwith the help of adversarial examples. In the absence of adversarial attacks,\nMWD networks match the performance of sigmoid networks, and have accuracy only\nslightly below that of ReLU networks.","url_abs":"http://arxiv.org/abs/1902.01208v1","url_pdf":"http://arxiv.org/pdf/1902.01208v1.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":"a-new-family-of-neural-networks-provably","repo_url":"https://github.com/rakshit-agrawal/mwd_nets","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[],"methods":[{"method_slug":"relu","method_name":"ReLU"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers","held":null,"reason":"arxiv_id_not_in_atlas","checked_against":{"atlas_ids":549449,"atlas_ids_sha256":"afd72399dafd9906e9062927f70655952385a3d75488558e1f0484e950ab09b7","read_at":"2026-10-05T01:46:21Z"},"by_repo":[{"repo":"rakshit-agrawal/mwd_nets","kind":"official","samples":0,"ran":0,"instrument":0,"unverified":0}],"by_repo_is":"one row per repository in the archive's code links, Syntology's graph links or the samples; samples 0 means no sample is linked to this paper, a statement about Syntology's coverage, not about the repository: with harvested_for_other_papers present, Syntology harvested the repository for that many other papers, and without it, Syntology harvested nothing from it; placed_by_identical_code counts samples whose link records no repository, placed here because identical code was harvested from this repository for another paper; ran counts executions on a synthesized input and instrument counts failures of Syntology's instrument, not of the code"},"code_links_note":null,"arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"94516806f936a029c72a10431bdbd0ace53dd326709e483ffe362fba53030974","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}