[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-119111-en":3,"doc-seo-119111-105":30,"detail-sidebar-cat-0-en-105":84},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},119111,2336464648322,"Aria","https://ap-avatar.wpscdn.com/avatar/2200025388227c56fec?_k=1778556882303663488",6,"Technology","AIJack - Security and Privacy Risk Simulator for Machine Learning - Open-Source Library","AIJack is an open-source library built to assess security and privacy risks in the training and deployment of machine learning models. With ML accelerating across research and business, threats such as training data theft and model manipulation require systematic evaluation. The library unifies a range of attack and defense methods behind a single API, enabling repeatable experiments that combine techniques. Designed for PyTorch and scikit-learn, AIJack supports practical risk simulation using accessible code.","arXiv :2312 . 17667v2 [ cs .LG] 8 Apr 2024  \nAIJack: Let’s Hijack AI!  \nSecurity and Privacy Risk Simulator for Machine Learning  \nHideaki Takahashi (Koukyosyumei)  \n[koukyosyumei@hotmaial. com](koukyosyumei@hotmaial. com)  \nAbstract  \nThis paper introduces AIJack, an open-source library designed to assess security and privacy risks associated with the training and deployment of machine learning models. Amid the growing interest in big data and AI, machine learning research and business advancements are accelerating. However, recent studies reveal potential threats, such as the theft of training data and the manipulation of models by malicious attackers. Therefore, a comprehensive understanding of machine learning’s security and privacy vulnerabilities is crucial for the safe integration of machine learning into real-world products. AIJack aims to address this need by providing a library with various attack and defense methods through a uni􀀌ed API. The library is publicly available on GitHub ([https://github.com/Koukyosyumei/AIJack](https://github.com/Koukyosyumei/AIJack) ).  \n1 Introduction  \nMachine learning (ML) has become a foundational component of diverse applications, spanning image recognition to natural language processing. As these technologies proliferate, comprehending and addressing security risks associated with ML models becomes imperative.  \nWhile ML models continuously enhance accuracy, attackers can signi􀀌cantly diminish it by introducing arti􀀌cially created data. Evasion Attacks [1, 2], using specialized algorithms, can induce model malfunctions. Consider a road sign classi􀀌cation model; adding imperceptible noise to a ”speed limit 30km” sign may misclassify it as ”speed limit 60km.” Those perturbed samples are often called adversarial examples. Another attack, Poisoning Attack [3], injects contaminated data during training, lowering model accuracy.  \nVarious strategies have been proposed to counter such attacks. Certi􀀌ed robustness [4 , 5], a technique to formally guarantee that adversarial examples cannot lead to undesirable predictions, has gained attention. Another approach is debugging machine learning models [6, 7 , 8], which aims to identify inputs causing unexpected behaviors.  \nML also poses privacy challenges. Collecting large amounts of data for training can infringe on privacy, leading to data breaches. Model Inversion Attacks [9 , 10] reconstruct training data from pre-trained models, threatening sensitive information. Similar attacks, like Membership Inference Attacks [11], aim to determine if a data point is part of the model’s training data.  \nPrivacy protection techniques, including di􀀋erential privacy [12], k-anonymity [13], homomorphic encryption [14], and distributed learning [15 , 16], have been proposed. Di􀀋erential privacy prevents individual data inference, while homomorphic encryption enables arithmetic operations on encrypted data. Distributed methods like federated learning (FL) [15] facilitate collaborative learning among data owners without violating data privacy.  \nThus, assessing ML models’ security and privacy risks and evaluating countermeasure e􀀋ectiveness is crucial. To simplify such simulations, we propose the open-source software, AIJack, o􀀋ering various attack and defense methods. AIJack enables the experimentation of numerous combinations of attacks and defenses with simple code. Built on PyTorch [17] and scikit-learn [18], users can easily incorporate AIJack into existing code.  \n2 Package Design  \nAIJack is designed with the following principles:  \nType  \nSubcategory  \nMethod  \nFL  \nHorizontal  \nFedAVG [15], FedProx [19], FedMD [20], FedGEMS [21], DSFL [22], MOON [23], FedEXP [24]  \nFL  \nVertical SplitNN [16], SecureBoost [25]  \n\n| Attack | Model Inversion | MI-FACE [9], DLG [10], iDLG [26], GS [27], CPL [28], GradInversion [29], GAN Attack [30] |  |\n| --- | --- | --- | --- |\n| Attack | Label Leakage | Norm Attack [31] |  |\n| Attack | Poisoning | History Attack [32], ","cbCaimJ1yYJlcIu8","https://ap.wps.com/l/cbCaimJ1yYJlcIu8","pdf",276696,1,16,"English","en",105,"# Introduction\n# Package Design","[{\"question\":\"How does AIJack help users run simulations in practice?\",\"answer\":\"AIJack offers a flexible, unified API and supports PyTorch and scikit-learn models, letting users combine attacks and defenses with relatively simple code for experimentation.\"}]","AIJack - Security and Privacy Risk Simulator for Machine Learning - Open-Source Library | PDF",1785722436,40,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":79,"head_meta":81,"extra_data":83,"updated_unix":28},"aijack-security-and-privacy-risk-simulator-for-machine-learning-open-source-library","",{"@graph":36,"@context":78},[37,54,69],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/technology/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/aijack-security-and-privacy-risk-simulator-for-machine-learning-open-source-library/119111/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04","2026-08-03",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72],{"name":73,"@type":74,"acceptedAnswer":75},"How does AIJack help users run simulations in practice?","Question",{"text":76,"@type":77},"AIJack offers a flexible, unified API and supports PyTorch and scikit-learn models, letting users combine attacks and defenses with relatively simple code for experimentation.","Answer","https://schema.org",{"og:url":52,"og:type":80,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":82,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":85},[86,90,94,98,103,106,110,115,120,123,127],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":87,"show_sort_weight":88,"slug":89},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":91,"show_sort_weight":92,"slug":93},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Exam",70,"exam",{"id":99,"doc_module":4,"doc_module_name":46,"category_name":100,"show_sort_weight":101,"slug":102},5,"Comic",60,"comic",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":104,"slug":105},50,"technology",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":29,"slug":109},7,"Healthcare","healthcare",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},8,"Research & Report",30,"research-report",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},9,"Religion & Spirituality",20,"religion-spirituality",{"id":118,"doc_module":4,"doc_module_name":46,"category_name":121,"show_sort_weight":118,"slug":122},"World Cup","world-cup",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":124,"slug":126},10,"Lifestyle","lifestyle",{"id":128,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":99,"slug":130},19,"General","general"]