Search

"This is the pre-peer reviewed version of the following article: Banks D, Gallego V, Naveiro R, Ríos Insua D. Adversarial risk analysis: An overview. WIREs Comput Stat. 2020;e1530. https://doi.org/10.1002/wics.153016, which has been published in final form at https://onlinelibrary.wiley.com/doi/epdf/10.1002/wics.1530. This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Use of Self-Archived Versions.", Adversarial risk analysis (ARA) is a relatively new area of research that informs decision-making when facing intelligent opponents and uncertain outcomes. It is a decision-theoretic alternative to game theory. ARA enables an analyst to express her Bayesian beliefs about an opponent's utilities, capabilities, probabilities, and the type of strategic calculations that the opponent is using to make his decision. Within that framework, the analyst then solves the problem from the perspective of the opponent. This calculation produces a distribution over the actions of the opponent that permits the analyst to maximize her expected utility. This review covers conceptual, modeling, computational, and applied issues in ARA as well as interesting open research issues., Unión Europea. Horizonte 2020, Ministerio de Ciencia e Innovación (MICINN)/FEDER, Ministerio de Ciencia e Innovación (MICCIN), National Science Foundation, Depto. de Estadística e Investigación Operativa, Fac. de Ciencias Matemáticas, FALSE, pub

Machine learning and computational processing have advanced such that automated driving systems (ADSs) are no longer a distant reality. Many automobile manufacturers have developed prototypes; however, there exist numerous decision support issues requiring resolution to ensure mass ADS adoption. In the coming decades, it is likely that production ADSs will only be partially autonomous. Such ADSs operate within predetermined conditions and require driver intervention when they are violated. Since forecasts of their 20-year market penetration are relatively low, ADSs will likely operate in heterogeneous traffic characterized by vehicles of varying autonomy levels. Under these conditions, effective decision support must consider intangible, subjective, and emotional factors as well as influences of human cognition; otherwise, the ADS risks driver distrust and unsatisfactory performance based on an incomplete understanding of its environment. We survey the literature relevant to these issues, identify open problems, and propose research directions for their resolution.