{"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/foolbox-a-python-toolbox-to-benchmark-the","title":"Foolbox: A Python toolbox to benchmark the robustness of machine learning models","arxiv_id":"1707.04131","date":"2017-07-13","proceeding":null,"authors":["Jonas Rauber","Wieland Brendel","Matthias Bethge"],"abstract":"Even todays most advanced machine learning models are easily fooled by almost\nimperceptible perturbations of their inputs. Foolbox is a new Python package to\ngenerate such adversarial perturbations and to quantify and compare the\nrobustness of machine learning models. It is build around the idea that the\nmost comparable robustness measure is the minimum perturbation needed to craft\nan adversarial example. To this end, Foolbox provides reference implementations\nof most published adversarial attack methods alongside some new ones, all of\nwhich perform internal hyperparameter tuning to find the minimum adversarial\nperturbation. Additionally, Foolbox interfaces with most popular deep learning\nframeworks such as PyTorch, Keras, TensorFlow, Theano and MXNet and allows\ndifferent adversarial criteria such as targeted misclassification and top-k\nmisclassification as well as different distance measures. The code is licensed\nunder the MIT license and is openly available at\nhttps://github.com/bethgelab/foolbox . The most up-to-date documentation can be\nfound at http://foolbox.readthedocs.io .","url_abs":"http://arxiv.org/abs/1707.04131v3","url_pdf":"http://arxiv.org/pdf/1707.04131v3.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":"foolbox-a-python-toolbox-to-benchmark-the","repo_url":"https://github.com/bethgelab/foolbox","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"foolbox-a-python-toolbox-to-benchmark-the","repo_url":"https://github.com/EvgeniaAR/foolbox","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"foolbox-a-python-toolbox-to-benchmark-the","repo_url":"https://github.com/alvarorobledo/fyp-foolbox","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"foolbox-a-python-toolbox-to-benchmark-the","repo_url":"https://github.com/bbwang1030/half-kfn","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"jax","reach":null},{"paper_slug":"foolbox-a-python-toolbox-to-benchmark-the","repo_url":"https://github.com/deqncho2/foolbox-adversarial","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"foolbox-a-python-toolbox-to-benchmark-the","repo_url":"https://github.com/pralab/secml","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"foolbox-a-python-toolbox-to-benchmark-the","repo_url":"https://gitlab.com/secml/secml","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"adversarial-attack","task_name":"Adversarial Attack"},{"task_slug":"machine-learning","task_name":"BIG-bench Machine Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1707.04131","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}