Election control problems are concerned with affecting the result of an election by modifying the structure of the election. Such election modifications could be either introducing some new candidates or voters or removing some existing candidates or voters from the election or partitioning candidates or voters [2, 27, 32, 42, 56, 57, 34, 35, 62]. We focus on the computational complexity of election control by adding and deleting candidates (that is, candidate control), for the case where the election involves only a few voters. From the viewpoint of computational complexity, voter control with few voters has not received sufficient study. We focus on very simple, practical voting rules such as Plurality, Veto, andt-Approval, but discuss several more involved rules as well. To analyze the effect of allowing only a small number of voters, we use the formal tools of parameterized complexity theory [21, 23, 38, 60]. From the viewpoint of classical complexity theory, most of the candidate control problems for most of the typically studied voting rules are NPhard. Indeed, candidate control problems are NPhard even for the Plurality rule; nonetheless, there are some natural examples of candidate control problems with polynomialtime algorithms. It turns out that for the case of elections with few voters, that is, for control problems parameterized by the number of voters, the computational complexity landscape of candidate control is much more varied and sometimes quite surprising.

Acquiring your first language is an incredible feat and not easily duplicated. Learning to communicate using nothing but a few pictureless books, a corpus, would likely be impossible even for humans. Nevertheless, this is the dominating approach in most natural language processing today. As an alternative, we propose the use of situated interactions between agents as a driving force for communication, and the framework of Deep Recurrent Q-Networks for evolving a shared language grounded in the provided environment. We task the agents with interactive image search in the form of the game Guess Who?. The images from the game provide a non trivial environment for the agents to discuss and a natural grounding for the concepts they decide to encode in their communication. Our experiments show that the agents learn not only to encode physical concepts in their words, i.e. grounding, but also that the agents learn to hold a multi-step dialogue remembering the state of the dialogue from step to step.

This paper analyses the DeGroot-Friedkin model for evolution of the individuals' social powers in a social network when the network topology varies dynamically (described by dynamic relative interaction matrices). The DeGroot-Friedkin model describes how individual social power (self-appraisal, self-weight) evolves as a network of individuals discuss a sequence of issues. We seek to study dynamically changing relative interactions because interactions may change depending on the issue being discussed. In order to explore the problem in detail, two different cases of issue-dependent network topologies are studied. First, if the topology varies between issues in a periodic manner, it is shown that the individuals' self-appraisals admit a periodic solution. Second, if the topology changes arbitrarily, under the assumption that each relative interaction matrix is doubly stochastic and irreducible, the individuals' self-appraisals asymptotically converge to a unique non-trivial equilibrium.

When we say "I know why he was late", we know not only the fact that he was late, but also an explanation of this fact. We propose a logical framework of "knowing why" inspired by the existing formal studies on why-questions, scientific explanation, and justification logic. We introduce the Ky_i operator into the language of epistemic logic to express "agent i knows why phi" and propose a Kripke-style semantics of such expressions in terms of knowing an explanation of phi. We obtain two sound and complete axiomatizations w.r.t. two different model classes depending on different assumptions about introspection.

Model checking of strategic ability under imperfect information is known to be hard. The complexity results range from NP-completeness to undecidability, depending on the precise setup of the problem. No less importantly, fixpoint equivalences do not generally hold for imperfect information strategies, which seriously hampers incremental synthesis of winning strategies. In this paper, we propose translations of ATLir formulae that provide lower and upper bounds for their truth values, and are cheaper to verify than the original specifications. That is, if the expression is verified as true then the corresponding formula of ATLir should also hold in the given model. We begin by showing where the straightforward approach does not work. Then, we propose how it can be modified to obtain guaranteed lower bounds. To this end, we alter the next-step operator in such a way that traversing one's indistinguishability relation is seen as atomic activity. Most interestingly, the lower approximation is provided by a fixpoint expression that uses a nonstandard variant of the next-step ability operator. We show the correctness of the translations, establish their computational complexity, and validate the approach by experiments with a scalable scenario of Bridge play.

Reality and the digital world are coming together. Cheap sensors, combined with the Internet of Things and big data processing, allow a company to monitor equipment, product, environment, and more. The end result is instant reaction at the point of interaction. This is a radical change in how business operates, according to Suresh Acharya (pictured), head of JDA Labs at JDA Software Group Inc. "What is changing dramatically is the fact all of this is being digitized," said Acharya. He spoke to John Furrier (@furrier), host of theCUBE, SiliconANGLE's mobile live-streaming studio, at the South by Southwest conference in Austin, TX. (*Disclosure below).

In the past year or so, a new marketing channel has emerged around bots and intelligent agents. This includes voice-based intelligent agents like Google Home's Assistant or Amazon's Alexa, and chatbots that interact largely through text conversations. Marketers are beginning to plan their conversational strategies and logic for this channel. But another channel -- maybe it should be considered a sub-channel -- is about to emerge. It's when the bot or agent, fulfilling the needs of the human user, is interacting with another bot or agent instead of searching the web, or a knowledge base, or a profile.

Reinforcement learning is an area of machine learning inspired by behaviorist psychology, concerned with how software agents ought to take actions in an environment so as to maximize some notion of cumulative reward. The problem, due to its generality, is studied in many other disciplines, such as game theory, control theory, operations research, information theory, simulation-based optimization, multi-agent systems, swarm intelligence, statistics, and genetic algorithms. In the operations research and control literature, the field where reinforcement learning methods are studied is called approximate dynamic programming. The problem has been studied in the theory of optimal control, though most studies are concerned with the existence of optimal solutions and their characterization, and not with the learning or approximation aspects. In economics and game theory, reinforcement learning may be used to explain how equilibrium may arise under bounded rationality.

In a new paper, researchers from the University of Missouri, MIT CISR and the London School of Economics analyze Swedish bank SEB's successful implementation of IPsoft's cognitive agent, Amelia. Professor Mary Lacity, Professor Leslie Willcocks and Associate Researcher Andrew Craig explore the benefits gained by early adopters of cognitive technologies. "Organizations still considering the adoption of virtual agents can learn valuable insights from SEB," the authors write. "A major lesson for future organizational adopters: If you just think of (Cognitive Virtual Agents) as a staff reduction tool, you will miss the competitive advantage and transformative potential of the technology." Please fill out the form below to download the report.

It seemed like such a simple thing. Yet another company had suffered a data breach, leaking its customers' private information all over the interwebs. You've probably been subject to such a leak yourself. This one was a bit different because it involved robots. Okay, nothing so dramatic as clanking mechanical men, but "bots" in the sense of automated software agents that converse with actual humans.