[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-121969-en":3,"doc-seo-121969-105":30,"detail-sidebar-cat-0-en-105":91},{"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":4,"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},121969,4398048950312,"Violet","https://ap-avatar.wpscdn.com/avatar/400002538284de19e3c?_k=1778320343897328908",8,"Research & Report","Tensor Networks for Explainable Machine Learning in Cybersecurity - Research","Tensor Networks for Explainable Machine Learning in Cybersecurity presents an unsupervised anomaly-detection clustering approach grounded in Matrix Product States (MPS). The work demonstrates that MPS-based models can match or rival deep learning baselines such as autoencoders and GANs on performance while offering substantially improved interpretability. The method enables feature-wise probabilities, Von Neumann entropy, and mutual information to support transparent, decision-rationale explanations in adversary-generated threat intelligence, aiding anomaly classification and cybersecurity analytics.","Tensor Networks for Explainable Machine Learning in Cybersecurity  \narXiv :2401 .00867v4 [ cs .LG] 25 Apr 2025  \nBorja Aizpurua, 1, 2 Samuel Palmer,3 and Rom´an Or´us 1, 4, 5  \n1 Multiverse Computing, Paseo de Miram´on 170, E-20014 San Sebasti´an, Spain  \n2 Department of Basic Sciences, Tecnun - University of Navarra, E-20018 San Sebasti´an, Spain  \n3 Multiverse Computing, Spadina Ave . , Toronto, ON M5T 2C2, Canada  \n4 Donostia International Physics Center, Paseo Manuel de Lardizabal 4, E-20018 San Sebasti´an, Spain  \n5 Ikerbasque Foundation for Science, Maria Diaz de Haro 3, E-48013 Bilbao, Spain  \nIn this paper we show how tensor networks help in developing explainability of machine learning algorithms. Specifically, we develop an unsupervised clustering algorithm based on Matrix Product States (MPS) and apply it in the context of a real use-case of adversary-generated threat intelligence. Our investigation proves that MPS rival traditional deep learning models such as autoencoders and GANs in terms of performance, while providing much richer model interpretability. Our approach naturally facilitates the extraction of feature-wise probabilities, Von Neumann Entropy, and mutual information, offering a compelling narrative for classification of anomalies and fostering an unprecedented level of transparency and interpretability, something fundamental to understand the rationale behind artificial intelligence decisions.  \nI. INTRODUCTION  \nExplainable Artificial Intelligence (XAI) emerges as a cornerstone in the advancement of AI, shedding light on the often opaque intricacies of algorithmic decisionmaking [1] . It strives to render machine learning models that are not only robust and precise but also transparent and comprehensible to human sight. The impetus for XAI is twofold: 1) it cultivates trust and supports robust decision-making by providing clear explanations for outcomes, and 2) it ensures compliance with increasingly stringent transparency regulations. XAI achieves these goals through specific techniques such as LIME (Local Interpretable Model-agnostic Explanations) [2] and SHAP (SHapley Additive exPlanations) [3], where the challenge lies in the tradeoff between model’s accuracy and explainability.  \nDeep learning methods, including neural networks, autoencoders [4] and GANs [5], excel at detecting intricate data patterns but often act as “black boxes”. Their complex architectures deliver powerful performance yet hinder the visibility of the decision-making process, posing a challenge to the increasing imperative of interpretability in machine learning. After all, one should be able to explain why an algorithm is providing a particular answer, and not a different one. In the search for a solution for this concern, the machine learning community has grown interest in alternative architectures that could make the work.  \nIn this work, we showcase the capabilities of Tensor Networks (TN) [6] to implement explainable machine learning. While other authors have explored similar trends recently [7], we go one step beyond and validate our approach with a real use-case in cybersecurity analytics.  \nTNs are a powerful mathematical framework used to represent high-dimensional data efficiently. They factorize vectors and operators in high-dimensional vector spaces into a network of lower-dimensional tensors, enabling to capture complex multi-partite correlations  \nwhile managing the curse of dimensionality. As such, TNs have a large number of applications in the simulation of complex quantum systems [8] . In the realm of machine learning, TNs have been traditionally considered in two approaches: 1) the tensorization of traditional machine learning models for enhanced computation, as demonstrated by recent research [9], and 2) the creation of intrinsic tensor network-based models, based on, e.g., Matrix Product States (MPS) [6, 10] . MPS, in particular, stands out as an efficient and transformative approach in explainable AI, di","cbCainSM28Zal55u","https://ap.wps.com/l/cbCainSM28Zal55u","pdf",942471,1,12,"English","en",105,"# Introduction\n## Explainable Artificial Intelligence and its motivation\n## Tensor Networks and Matrix Product States\n# Methodology\n## Tensor networks and anomaly detection with MPS\n# Performance and Interpretability\n## Comparative results in threat intelligence\n## Feature probabilities and information-theoretic measures\n# Conclusion","[{\"question\":\"What explainability capabilities does the MPS-based clustering algorithm provide?\",\"answer\":\"The approach supports interpretability through feature-wise probabilities, Von Neumann entropy, and mutual information, enabling transparent reasoning for anomaly classification.\"},{\"question\":\"How does the proposed method compare with traditional deep learning models?\",\"answer\":\"The paper reports that MPS can rival deep learning models such as autoencoders and GANs in performance while providing richer model interpretability.\"},{\"question\":\"Why is the adversary-generated threat intelligence use case central to the study?\",\"answer\":\"It tests the algorithm in cybersecurity analytics by modeling normal behavior and identifying deviations that may indicate sophisticated cyber-attacks, supporting anomaly detection and early response.\"}]","Tensor Networks for Explainable Machine Learning in Cybersecurity - Research | PDF",1785808074,30,{"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":86,"head_meta":88,"extra_data":90,"updated_unix":28},"tensor-networks-for-explainable-machine-learning-in-cybersecurity-research","",{"@graph":36,"@context":85},[37,54,68],{"@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/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/tensor-networks-for-explainable-machine-learning-in-cybersecurity-research/121969/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What explainability capabilities does the MPS-based clustering algorithm provide?","Question",{"text":75,"@type":76},"The approach supports interpretability through feature-wise probabilities, Von Neumann entropy, and mutual information, enabling transparent reasoning for anomaly classification.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the proposed method compare with traditional deep learning models?",{"text":80,"@type":76},"The paper reports that MPS can rival deep learning models such as autoencoders and GANs in performance while providing richer model interpretability.",{"name":82,"@type":73,"acceptedAnswer":83},"Why is the adversary-generated threat intelligence use case central to the study?",{"text":84,"@type":76},"It tests the algorithm in cybersecurity analytics by modeling normal behavior and identifying deviations that may indicate sophisticated cyber-attacks, supporting anomaly detection and early response.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,122,127,130,134],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":29,"slug":121},"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]