Adversarial Action Research Articles

EFraudCom: An E-commerce Fraud Detection System via Competitive Graph Neural Networks

With the development of e-commerce, fraud behaviors have been becoming one of the biggest threats to the e-commerce business. Fraud behaviors seriously damage the ranking system of e-commerce platforms and adversely influence the shopping experience of users. It is of great practical value to detect fraud behaviors on e-commerce platforms. However, the task is non-trivial, since the adversarial action taken by fraudsters. Existing fraud detection systems used in the e-commerce industry easily suffer from performance decay and can not adapt to the upgrade of fraud patterns, as they take already known fraud behaviors as supervision information to detect other suspicious behaviors. In this article, we propose a competitive graph neural networks (CGNN)-based fraud detection system (eFraudCom) to detect fraud behaviors at one of the largest e-commerce platforms, “Taobao” 1 . In the eFraudCom system, (1) the competitive graph neural networks (CGNN) as the core part of eFraudCom can classify behaviors of users directly by modeling the distributions of normal and fraud behaviors separately; (2) some normal behaviors will be utilized as weak supervision information to guide the CGNN to build the profile for normal behaviors that are more stable than fraud behaviors. The algorithm dependency on fraud behaviors will be eliminated, which enables eFraudCom to detect fraud behaviors in presence of the new fraud patterns; (3) the mutual information regularization term can maximize the separability between normal and fraud behaviors to further improve CGNN. eFraudCom is implemented into a prototype system and the performance of the system is evaluated by extensive experiments. The experiments on two Taobao and two public datasets demonstrate that the proposed deep framework CGNN is superior to other baselines in detecting fraud behaviors. A case study on Taobao datasets verifies that CGNN is still robust when the fraud patterns have been upgraded.

Self-Adaptive Software Systems in Contested and Resource-Constrained Environments: Overview and Challenges

Self-adaptive approaches are a promising to address the dynamic and uncertain nature of the environments where today’s complex systems operate. In particular, systems operating in military environments, during crises or under unexpected conditions need to address critical concerns related to resource sparsity and the unstable and uncertain nature of exchanged information. Despite a plethora of self-adaptive and autonomic approaches proposed in the last decade, very few have been designed for or evaluated in contested environments , where adversarial action in the communications domain leads to stale or incomplete information, or in resource-constrained environments, where resources are either limited or required by a large number of components. These conditions are where self-adaptability to aid human operators is needed the most. To better understand self-adaptation in contested and resource-constrained environments, we conducted a systematic literature review of publications over the last decade. We conduct our review through the lens of a military environment, where contention, both physical and from a resource, be it computational or communication based is at its peak. We followed the systematic literature review methodology and analysed 238 primary studies. We identified that the most frequent application domains are those where failures are frequent and costly, namely, cloud computing, web services and applications, and servers. Despite this, less than 3% of the papers considered constrained resources and stale or incomplete information and a significant focus was on developing centralised solutions instead of distributed ones throughout all papers. Very few papers (4.6%) considered environments where the information about the system components was not readily available. No papers evaluated the systems running in contested and resource-constrained environments. We present an analysis of the self-adaptive systems that consider incomplete or stale information and constrained resources, discuss their limitations and identify future areas of research. Critical research gaps include the lack of evaluation of self-adaptive approaches, including the lack of standards or formalisms to allow for the comparison of various approaches. In addition, there is a need to consider more self-* properties and non-functional requirements in order to make sure the designed system is resilient. This paper presents our review findings in detail, examines how self-adaptation happens in contested and resource-constrained environments, and discusses the identified research gaps.

Tunable Measures for Information Leakage and Applications to Privacy-Utility Tradeoffs

We introduce a tunable measure for information leakage called maximal alpha-leakage. This measure quantifies the maximal gain of an adversary in inferring any (potentially random) function of a dataset from a release of the data. The inferential capability of the adversary is, in turn, quantified by a class of adversarial loss functions that we introduce as $\alpha$-loss, $\alpha\in[1,\infty]$. The choice of $\alpha$ determines the specific adversarial action and ranges from refining a belief (about any function of the data) for $\alpha=1$ to guessing the most likely value for $\alpha=\infty$ while refining the $\alpha^{th}$ moment of the belief for $\alpha$ in between. Maximal alpha-leakage then quantifies the adversarial gain under $\alpha$-loss over all possible functions of the data. In particular, for the extremal values of $\alpha=1$ and $\alpha=\infty$, maximal alpha-leakage simplifies to mutual information and maximal leakage, respectively. For $\alpha\in(1,\infty)$ this measure is shown to be the Arimoto channel capacity of order $\alpha$. We show that maximal alpha-leakage satisfies data processing inequalities and a sub-additivity property thereby allowing for a weak composition result. Building upon these properties, we use maximal alpha-leakage as the privacy measure and study the problem of data publishing with privacy guarantees, wherein the utility of the released data is ensured via a hard distortion constraint. Unlike average distortion, hard distortion provides a deterministic guarantee of fidelity. We show that under a hard distortion constraint, for $\alpha>1$ the optimal mechanism is independent of $\alpha$, and therefore, the resulting optimal tradeoff is the same for all values of $\alpha>1$. Finally, the tunability of maximal alpha-leakage as a privacy measure is also illustrated for binary data with average Hamming distortion as the utility measure.

Psychology and Its Response to Major Human Rights Abuses

Abstract. Australian immigration detention has been criticized both domestically and internationally for the harm that it creates and promotes and for violating human rights and international law. Psychologists have worked within centers and have thus been central to their operation, but have also long called for reform of these policies. Despite this and despite broader criticism from all corners of Australian society, the government has continued to consolidate power in relation to the administration of these policies and has actively attempted to shut down dissent. How should Psychologists respond? This article will argue that current approaches are inadequate and more adversarial action is needed. Supporting such an approach, social movement theory will be introduced and applied to examine how it may inform future action. Psychologists have an obligation to protect human rights and health, and while more adversarial action may not typically fit in traditional repertoires, there are few other professionals who are better skilled to begin to deal with these questions. In light of this, Psychologists in Australia and across the globe should carefully consider their roles in social change and whether they can do more in the face of major human rights abuses.

WHAT THE PROFESSION OF CYBERSECURITY NEEDS TO KNOW AND DO

ABSTRACTPresently, 71% of annual losses are due to failures in the physical and human attack domains, while electronic breaches account for roughly 29%. While the lowest percentage of losses (29%) falls into the area of the classic technology-based attacks, unfortunately these are often the only kind of attacks factored into an organization’s cybersecurity planning. Surprisingly, in most organizations, human or physical types of threats are simply not part of traditional cyberdefense thinking. Most active cyberdefense solutions do not consider embodying integrated and well-defined behavioral controls into the cybersecurity process. And as a result, well executed attacks against the non-electronic attack surface are almost certain to succeed. We argue that the profession must find ways to ensure that the real-world practice of cybersecurity involves the creation and adoption of a complete, correct, and highly effective set of well-defined and commonly accepted controls; ones that are capable of closing off every feasible type of adversarial action. To be completely effective, the solution must amalgamate all of the essential concepts of cyberdefense into a single unifying practice model, one that has real-world currency. Professional societies help to serve as the developers and sanctioners of the fundamental ideas in their respective fields and the creation of the CSEC2017 document provides an authoritative statement of the elements of the field of cybersecurity for a broad array of practitioners. This paper discusses the CSEC2017 thought model and outlines the eight knowledge areas specified for the discipline to represent the complete body of knowledge within the field.

Adversarial Action Prediction Networks.

Different from after-the-fact action recognition, action prediction task requires action labels to be predicted from partially observed videos containing incomplete action executions. It is challenging because these partial videos have insufficient discriminative information, and their temporal structure is damaged. We study this problem in this paper, and propose an efficient and powerful deep network for learning representative and discriminative features for action prediction. Our approach exploits abundant sequential context information in full videos to enrich the feature representations of partial videos. This information is encoded in latent representations using a variational autoencoder (VAE), which are encouraged to be progress-invariant. Decoding such latent representations using another VAE, we can reconstruct missing information in the features extracted from partial videos. An adversarial learning scheme is adopted to differentiate the reconstructed features from the features directly extracted from full videos in order to well align their distributions. A multi-class classifier is also used to encourage the features to be discriminative. Our network jointly learns features and classifiers, and generates the features particularly optimized for action prediction. Extensive experimental results on UCF101, Sports-1M and BIT datasets demonstrate that our approach remarkably outperforms state-of-the-art methods, and shows significant speedup over these methods. Results also show that actions differ in their prediction characteristics; some actions can be correctly predicted even though only the beginning 10% portion of videos is observed.

THE EFFECT OF THE NEW ZEALAND ACCIDENT COMPENSATION LEGISLATION ON MEDICAL MALPRACTICE

The New Zealand Accident Compensation Act 1972 and its amending Act of 1982 have substantially altered the attitudes of the law, the medical profession and the public, to personal injury by accident. The element of fault is removed and the uncertainty of the outcome of a claim under adversarial court action no longer exists. Rehabilitation assistance on recovery from injury as well as accident prevention measures are cited in the Act as statutory responsibilities of the New Zealand Accident Compensation Corporation. Mechanisms of dealing with medical malpractice, with illustrative examples from ophthalmic practice, are described. Most such incidents are described as medical misadventure.