Arms Race in Adversarial Malware Detection: A Survey

Malicious software (malware) is a major cyber threat that has to be tackled with Machine Learning (ML) techniques because millions of new malware examples are injected into cyberspace on a daily basis. However, ML is vulnerable to attacks known as adversarial examples. In this article, we survey and systematize the field of Adversarial Malware Detection (AMD) through the lens of a unified conceptual framework of assumptions, attacks, defenses, and security properties. This not only leads us to map attacks and defenses to partial order structures, but also allows us to clearly describe the attack-defense arms race in the AMD context. We draw a number of insights, including: knowing the defender’s feature set is critical to the success of transfer attacks; the effectiveness of practical evasion attacks largely depends on the attacker’s freedom in conducting manipulations in the problem space; knowing the attacker’s manipulation set is critical to the defender’s success; and the effectiveness of adversarial training depends on the defender’s capability in identifying the most powerful attack. We also discuss a number of future research directions.

The Role of Trustworthiness Facets for Developing Social Media Applications: A Structured Literature Review

This work reviews existing research about attributes, which are assessed by individuals to evaluate the trustworthiness of (i) software applications, (ii) organizations (e.g., service providers), and (iii) other individuals. As these parties are part of social media services, previous research has identified the need for users to assess their trustworthiness. Based on the trustworthiness assessment, users decide whether they want to interact with them and whether such interactions appear safe. The literature review encompasses 264 works from which so-called trustworthiness facets of 100 papers could be identified. In addition to an overview of trustworthiness facets, this work further introduces a guideline for software engineers on how to select appropriate trustworthiness facets during the analysis of the problem space for the development of specific social media applications. It is exemplified by the problem of “catfishing” in online dating.

MENTAL HEALTH ACCESS BACKGROUND AND FRAMEWORK FOR IMPROVEMENT IN THE DIGITAL ERA

Mental Health consists of a wide range of conditions affecting an individual’s behavior, mood, and thinking. A gap between providers and consumers exists due to the lack of resources available for online mental health access. Research shows limited information available for processing information and pursuing professional services to become healthy. The purpose of this literature review is to provide framework that contributes to Mental Health online access and improving digital therapy. Suggested approaches listed in addition to graphics that summarize the problem space and provide potential solutions.

The Power of Many: A Physarum Swarm Steiner Tree Algorithm

This paper presents a novel explore-and-fuse approach to solving a large array of problems that cannot be solved by traditional divide-and-conquer. This approach is inspired by Physarum, a unicellular slime mold capable of solving the traveling salesman and Steiner tree problems. Besides exhibiting individual intelligence, Physarum can also share information with other Physarum organisms through fusion. Inspired by the characteristics of Physarum, we spawn many Physarum organisms to explore the problem space in parallel, each gathering information and forming partial solutions pertaining to a local region of the problem space. When the organisms meet, they fuse and share information, eventually forming one organism which has a global view of the problem and can apply its intelligence to find an overall solution to the problem. We demonstrate this novel approach on the NP-hard Steiner tree problem, developing the Physarum Steiner Algorithm. This algorithm is of particular interest due to its ability to leverage parallel processing, avoid obstacles, and operate on various shapes and topological surfaces including the rectilinear grid.

Deep Reinforcement Learning For Trading—A Critical Survey

Deep reinforcement learning (DRL) has achieved significant results in many machine learning (ML) benchmarks. In this short survey, we provide an overview of DRL applied to trading on financial markets with the purpose of unravelling common structures used in the trading community using DRL, as well as discovering common issues and limitations of such approaches. We include also a short corpus summarization using Google Scholar. Moreover, we discuss how one can use hierarchy for dividing the problem space, as well as using model-based RL to learn a world model of the trading environment which can be used for prediction. In addition, multiple risk measures are defined and discussed, which not only provide a way of quantifying the performance of various algorithms, but they can also act as (dense) reward-shaping mechanisms for the agent. We discuss in detail the various state representations used for financial markets, which we consider critical for the success and efficiency of such DRL agents. The market in focus for this survey is the cryptocurrency market; the results of this survey are two-fold: firstly, to find the most promising directions for further research and secondly, to show how a lack of consistency in the community can significantly impede research and the development of DRL agents for trading.

Exploring the Problem Space of CO2 Emission Reductions from Academic Flying

CO2 emissions from aviation have been predicted to increase over the coming decades. Within the academic world, flying is often perceived to be a necessary prerequisite to being a successful researcher. Many Swedish universities have ambitious climate goals, but are simultaneously among the top emitters in the public sector. Reaching stated climate goals could feasibly be met through a combination of measures, including decreased flying. One way to address the challenge is to support behavioural interventions with the help of interactive visualizations of CO2 emissions from flying. Those few examples that exist in the research literature are generally directed towards management and are less applicable to universities, given the large autonomy researchers enjoy and their discretionary control of research project funds. This paper uses a design-oriented research approach to present an analysis of the problem space at the intersection of interactive visualizations using air travel data to reduce CO2 emissions from business air travel at our own university, KTH Royal Institute of Technology. Through a number of design experiments, evaluations and investigations, we have unearthed needs, challenges and opportunities for the creation of visualization tools to support more sustainable travel practices at universities and in other knowledge-intensive organisations.

Deep Reinforcement Learning For Trading - A Critical Survey

Deep reinforcement learning (DRL) has achieved significant results in many Machine Learning (ML) benchmarks. In this short survey we provide an overview of DRL applied to trading on financial markets, including a short meta-analysis using Google Scholar, with an emphasis on using hierarchy for dividing the problem space as well as using model-based RL to learn a world model of the trading environment which can be used for prediction. In addition, multiple risk measures are defined and discussed, which not only provide a way of quantifying the performance of various algorithms, but they can also act as (dense) reward-shaping mechanisms for the agent. We discuss in detail the various state representations used for financial markets, which we consider critical for the success and efficiency of such DRL agents. The market in focus for this survey is the cryptocurrency market.

From Social Gaze to Indirect Speech Constructions: How to Induce the Impression That Your Companion Robot Is a Conscious Creature

We implemented different modes of social gaze behavior in our companion robot, F-2, to evaluate the impression of the gaze behaviors on humans in three symmetric communicative situations: (a) the robot telling a story, (b) the person telling a story to the robot, and (c) both parties communicating about objects in the real world while solving a Tangram puzzle. In all the situations the robot localized the human’s eyes and directed its gaze between the human, the environment, and the object of interest in the problem space (if it existed). We examined the balance between different gaze directions as the novel key element to maintaining a feeling of social connection with the robot in humans. We extended the computer model of the robot in order to simulate realistic gaze behavior in the robot and create the impression of the robot changing its internal cognitive states. Other novel results include the implicit, rather than explicit, character of the robot gaze perception for many of our subjects and the role of individual differences, especially the level of emotional intelligence, in terms of human sensitivity to the robotic gaze. Therefore, in this study, we used an iterative approach, extending the applied cognitive architecture in order to simulate the balance between different behavioral reactions and to test it in the experiments. In such a way, we came to a description of the key behavioral cues that suggest to a person that the particular robot can be perceived as an emotional and even conscious creature.

Almost Risking It All

This paper provides an argument against understanding risk-taking in design education as something ideally in need of only being calculable and formalisable. Using the German sociologist Ulrich Beck’s theory on risktaking combined with the current discourse on design thinking, together with an analysis of a three week-long interdisciplinary design workshop, we analyse and discuss how risk-taking - as a general concept - in design education is an inherent element of the education itself. We argue, however, non-calculable risks, like human-centred design concerns, like desirability of use, ethics of technology, are an equally important part of a modern-day educational skillset as calculable risks. The aim is arguing for the prospect of interdisciplinary design-based education models as one way of embracing the non-calculable elements of a problem space.

The Covid-19 Learning Crisis as a Challenge and an Opportunity for Schools: An Evidence Review and Conceptual Synthesis of Research-Based Tools for Sustainable Change

This paper advances our understanding of how schools can become change agents capable of transforming local practice to address the challenges arising from the Covid-19 pandemic. It presents a novel application of cultural-historical activity theory to reinterpret evidence on widespread learning loss and increasing educational inequities resulting from the pandemic, and to identify scalable transformative learning opportunities through reframing the crisis as a double stimulation. By reviewing evidence of the emerging educational landscape, we first develop a picture of the new ‘problem space’ upon which schools must act. We develop a problem space map to serve as the first stimulus to articulate local challenges. Integrating this problem space with research on professional change, we identify conceptual tools to capture learning gaps and implement pedagogic interventions at scale, in order to enhance schools’ agency in directly addressing the crisis. These tools can act as the second stimulus, enabling educators to address local challenges. We conclude by discussing the Covid-19 educational crisis as a unique stimulus for professional learning and outline the potential for durable shifts in educational thinking and practice beyond the pandemic. We argue that this unprecedented historic disruption can be harnessed as a transformative professional learning opportunity. In particular, we consider how research on professional change offers local, scalable interventions and tools that can support educators in preventing the new insights from ‘slipping away’ post-pandemic. Utilising the notions of boundaries and tool-mediated professional change, we examine the ways in which this disruption generates opportunities to envision alternative futures for equitable learning in school.