Wanlei Zhou – författare

This book covers and makes four major contributions: 1) analyzing and surveying the pros and cons of current approaches for identifying rumor sources on complex networks; 2) proposing a novel approach to identify rumor sources in time-varying networks; 3) developing a fast approach to identify multiple rumor sources; 4) proposing a community-based method to overcome the scalability issue in this research area. These contributions enable rumor source identification to be applied effectively in real-world networks, and eventually diminish rumor damages, which the authors rigorously  illustrate in this book.

In the modern world, the ubiquity of networks has made us vulnerable to various risks. For instance, viruses propagate throughout the Internet and infect millions of computers. Misinformation spreads incredibly fast in online social networks, such as Facebook and Twitter. Infectious diseases, such as SARS, H1N1 or Ebola, have spread geographically and killed hundreds of thousands people. In essence, all of these situations can be modeled as a rumor spreading through a network, where the goal is to find the source of the rumor so as to control and prevent network risks. So far, extensive work has been done to develop new approaches to effectively identify rumor sources. However, current approaches still suffer from critical weaknesses. The most serious one is the complex spatiotemporal diffusion process of rumors in time-varying networks, which is the bottleneck of current approaches. The second problem lies in the expensively computational complexity of identifying multiple rumor sources. The third important issue is the huge scale of the underlying networks, which makes it difficult to develop efficient strategies to quickly and accurately identify rumor sources. These weaknesses prevent rumor source identification from being applied in a broader range of real-world applications. This book aims to analyze and address these issuesto make rumor source identification more effective and applicable in the real world.

The authors propose a novel reverse dissemination strategy to narrow down the scale of suspicious sources, which dramatically promotes the efficiency of their method. The authors then develop a Maximum-likelihood estimator, which can pin point the true source from the suspects with high accuracy. For the scalability issue in rumor source identification, the authors explore sensor techniques and develop a community structure based method. Then the authors take the advantage of the linear correlation between rumor spreading time and infection distance, and develop a fast method to locate the rumor diffusion source. Theoretical analysis proves the efficiency of the proposed method, and the experiment results verify the significant advantages of the proposed method in large-scale networks.

This book targets graduate and post-graduate students studying computer science and networking.  Researchers and professionals working in network security, propagation models and other related topics, will also be interested in this book.

A critical challenge in deep learning is the vulnerability of deep learning networks to security attacks from intelligent cyber adversaries. Even innocuous perturbations to the training data can be used to manipulate the behaviour of deep networks in unintended ways.  In this book, we review the latest developments in adversarial attack technologies in computer vision; natural language processing; and cybersecurity with regard to multidimensional, textual and image data, sequence data, and temporal data. In turn, we assess the robustness properties of deep learning networks to produce a taxonomy of adversarial examples that characterises the security of learning systems using game theoretical adversarial deep learning algorithms. The state-of-the-art in adversarial perturbation-based privacy protection mechanisms is also reviewed.

We propose new adversary types for game theoretical objectives in non-stationary computational learning environments. Proper quantificationof the hypothesis set in the decision problems of our research leads to various functional problems, oracular problems, sampling tasks, and optimization problems. We also address the defence mechanisms currently available for deep learning models deployed in real-world environments. The learning theories used in these defence mechanisms concern data representations, feature manipulations, misclassifications costs, sensitivity landscapes, distributional robustness, and complexity classes of the adversarial deep learning algorithms and their applications.

In closing, we propose future research directions in adversarial deep learning applications for resilient learning system design and review formalized learning assumptions concerning the attack surfaces and robustness characteristics of artificial intelligence applications so as to deconstruct the contemporary adversarial deep learning designs. Given its scope, the book will be of interest to Adversarial Machine Learning practitioners and Adversarial Artificial Intelligence researchers whose work involves the design and application of Adversarial Deep Learning.

This book focuses on differential privacy and its application with an emphasis on technical and application aspects. This book also presents the most recent research on differential privacy with a theory perspective. It provides an approachable strategy for researchers and engineers to implement differential privacy in real world applications.

Early chapters are focused on two major directions, differentially private data publishing and differentially private data analysis. Data publishing focuses on how to modify the original dataset or the queries with the guarantee of differential privacy. Privacy data analysis concentrates on how to modify the data analysis algorithm to satisfy differential privacy, while retaining a high mining accuracy. The authors also introduce several applications in real world applications, including recommender systems and location privacy