Sequential decision making, commonly formalized as Markov Decision Process (MDP) optimization, is a key challenge in artificial intelligence. Two key approaches to this problem are reinforcement learning (RL) and planning. This paper presents a survey of the integration of both fields, better known as model-based reinforcement learning. Model-based RL has two main steps. First, we systematically cover approaches to dynamics model learning, including challenges like dealing with stochasticity, uncertainty, partial observability, and temporal abstraction.
Catholijn M. Jonker
Artificially intelligent agents will deal with more morally sensitive situations as the field of AI progresses. Research efforts are made to regulate, design and build Artificial Moral Agents (AMAs) capable of making moral decisions. This research is highly multidisciplinary with each their own jargon and vision, and so far it is unclear whether a fully autonomous AMA can be achieved. To specify currently available solutions and structure an accessible discussion around them, we propose to apply Team Design Patterns (TDPs).
The commercial availability of robots and voice-operated smart devices such as Alexa or Google Home have some companies wondering whether they can replace some current human interactions by using these devices. One such area of interaction is at the reception desk. While both platforms can offer the necessary interaction features to take on the task of an automated receptionist, the question remains as to which platform actual visitors would prefer - body or no body?
Social simulations gain strength when agent behaviour can (1) represent human behaviour and (2) be explained in understandable terms. Agents with values and norms lead to simulation results that meet human needs for explanations, but have not been tested on their ability to reproduce human behaviour. This paper compares empirical data on human behaviour to simulated data on agents with values and norms in a psychological experiment on dividing money: the ultimatum game.
In this article, we introduce a new paradigm to achieve Pareto optimality in group decision-making processes: bottom-up approaches to Pareto optimality. It is based on the idea that, while resolving a conflict in a group, individuals may trust some members more than others; thus, they may be willing to cooperate and share more information with those members. Therefore, one can divide the group into subgroups where more cooperative mechanisms can be formed to reach Pareto optimal outcomes.
Since its origination in the mid-twentieth century, the area of Artificial Intelligence (AI) has undergone a number of developments. While the early interest in AI was mainly triggered by the desire to develop artifacts that show the same intelligent behavior as humans, nowadays scientists have realized that research in AI involves a multitude of separate challenges, besides the traditional goal to replicate human intelligence.