Early in 2020, a US Treasury Department watchdog warned the IRS that the agency needed to do more to protect against identity fraud. In recent data breaches, the report said, "much of the information the IRS uses to provide assurance of the taxpayers' identities may have been stolen." The pandemic soon underscored the danger. When the IRS launched a web page for taxpayers to enter bank details for stimulus checks, it verified users by asking for data such as a person's birth date and Social Security number. Some people logged on only to find fraudsters had got there first.

An artificial intelligence algorithm destroyed a seasoned US fighter pilot in a simulated dogfight last week, a result some expert observers say was to be expected. "I was not surprised by that outcome," Guy'Bus' Snodgrass, a former US Navy pilot and TOPGUN instructor, told Insider, arguing that the set-up of the engagement gave the AI an advantage. John "JV" Venable, a former US Air Force F-16 pilot, said the same. The Defense Advanced Research Projects Agency (DARPA) held the last round of its third and final AlphaDogfight competition Thursday, putting an AI system designed by Heron Systems against a human pilot in a "simulated within-visual-range air combat" situation. The AI algorithm, which had previously defeated other AI "pilots," achieved a flawless victory, winning five straight matches without the human pilot -- an experienced Air Force pilot and Weapons Instructor Course graduate with the call sign "Banger" -- ever scoring a hit.

The more AI agents are deployed in scenarios with possibly unexpected situations, the more they need to be flexible, adaptive, and creative in achieving the goal we have given them. Thus, a certain level of freedom to choose the best path to the goal is inherent in making AI robust and flexible enough. At the same time, however, the pervasive deployment of AI in our life, whether AI is autonomous or collaborating with humans, raises several ethical challenges. AI agents should be aware and follow appropriate ethical principles and should thus exhibit properties such as fairness or other virtues. These ethical principles should define the boundaries of AI's freedom and creativity. However, it is still a challenge to understand how to specify and reason with ethical boundaries in AI agents and how to combine them appropriately with subjective preferences and goal specifications. Some initial attempts employ either a data-driven example-based approach for both, or a symbolic rule-based approach for both. We envision a modular approach where any AI technique can be used for any of these essential ingredients in decision making or decision support systems, paired with a contextual approach to define their combination and relative weight. In a world where neither humans nor AI systems work in isolation, but are tightly interconnected, e.g., the Internet of Things, we also envision a compositional approach to building ethically bounded AI, where the ethical properties of each component can be fruitfully exploited to derive those of the overall system. In this paper we define and motivate the notion of ethically-bounded AI, we describe two concrete examples, and we outline some outstanding challenges.

Preference are central to decision making by both machines and humans. Representing, learning, and reasoning with preferences is an important area of study both within computer science and across the sciences. When working with preferences it is necessary to understand and compute the distance between sets of objects, e.g., the preferences of a user and a the descriptions of objects to be recommended. We present CPDist, a novel neural network to address the problem of learning to measure the distance between structured preference representations. We use the popular CP-net formalism to represent preferences and then leverage deep neural networks to learn a recently proposed metric function that is computationally hard to compute directly. CPDist is a novel metric learning approach based on the use of deep siamese networks which learn the Kendal Tau distance between partial orders that are induced by compact preference representations. We find that CPDist is able to learn the distance function with high accuracy and outperform existing approximation algorithms on both the regression and classification task using less computation time. Performance remains good even when CPDist is trained with only a small number of samples compared to the dimension of the solution space, indicating the network generalizes well.

Questions about the imminent release of the convicted rapist, John Warboys, continue to dominate many of the papers. The Daily Mail asks why wasn't he given a longer minimum sentence, while the Times wants the Parole Board to explain its decision to free him. His ex-wife tells the Daily Express he should never be let out and the Daily Mirror insists "this sex monster should still be behind bars." There may yet be further prosecutions, according to the Guardian. It says alleged victims are considering bringing fresh cases.

In shows like HBO's "Westworld" and AMC's "Humans," Hollywood pits robots, with artificial intelligence, against humans. Half a century ago, a science fiction film about a space mission planted the first seeds of doubt about just how the human race could coexist with man-made sentient beings. "Consider the fictional robot HAL in '2001: A Space Odyssey,' " said Ken Ford, a computer scientist and founder and CEO of the Florida Institute for Human and Machine Cognition, in Pensacola, which has won awards for its robotics innovations. HAL eventually turned on its master in that classic film, sending shivers down the spines of moviegoers everywhere. Some of that wariness about artificial intelligence still exists, but Ford said the fear is unwarranted, and in the case of fictional robots, misplaced.

In shows like HBO's "Westworld" and AMC's "Humans," Hollywood pits robots, with artificial intelligence, against humans. Half a century ago, a science fiction film about a space mission planted the first seeds of doubt about just how the human race could coexist with man-made sentient beings. "Consider the fictional robot HAL in '2001: A Space Odyssey,' " said Ken Ford, a computer scientist and founder and CEO of the Florida Institute for Human and Machine Cognition, in Pensacola, which has won awards for its robotics innovations. HAL eventually turned on its master in that classic film, sending shivers down the spines of moviegoers everywhere. Some of that wariness about artificial intelligence still exists, but Ford said the fear is unwarranted, and in the case of fictional robots, misplaced.

m CP-nets are an expressive and intuitive formalism based on CP-nets to reason about preferences of groups of agents. The dominance semantics of mCP-nets is based on the concept of voting, and different voting schemes give rise to different dominance semantics for the group. Unlike CP-nets, which received an extensive complexity analysis, m CP-nets, as reported multiple times in the literature, lack a precise study of the voting tasks' complexity. Prior to this work, only a complexity analysis of brute-force algorithms for these tasks was available, and this analysis only gave EXPTIME upper bounds for most of those problems. In this paper, we start to fill this gap by carrying out a precise computational complexity analysis of voting tasks on acyclic binary polynomially connected m CP-nets whose constituents are standard CP-nets. Interestingly, all these problems actually belong to various levels of the polynomial hierarchy, and some of them even belong to PTIME or LOGSPACE. Furthermore, for most of these problems, we provide completeness results, which show tight lower bounds for problems that (up to date) did not have any explicit non-obvious lower bound.

In this challenge paper, we consider the importance of preferences in smart homes and assistive environments and discuss the potential application of models and algorithms developed within the computational preferences community. We suggest the value of future research collaborations.

Many real world applications require managing both system requirements and user preferences where the latter are usually provided in a qualitative way. We introduce a new approach to handle these two aspects, in an efficient way, respectively through Constraint Satisfaction Problems (CSPs) and CP-nets. In particular, we use Arc Consistency (AC) in order to reduce the search space needed when looking for the optimal outcome in an acyclic CP-net. More precisely, assuming that there are always some shared variables between the CP-net and the CSP, our approach works by first applying AC to the CSP and then update the CP-net with the remaining variables values. The resulting simplified CP-net will then be used to look for the best outcome. Experimental tests conducted on randomly generated problem instances clearly show the effect of AC on the size of the search space and the time needed to find the best outcome.