Catholijn M. Jonker

Toward AI Systems that Augment and Empower Humans by Understanding Us, our Society and the World Around Us

This report describes results from an initiative to organize a community of researchers and innovators around a research program that seeks to create AI technologies that empower humans and human society to vastly improve quality of life for all. The introduction describes the context and motivation for an initiative to organize the Humane AI community. Section 2 describes the Humane AI Vision to create the foundations for AI systems that empower people and society.

Towards Agent-based Models of Rumours in Organizations: A Social Practice Theory Approach

Rumour is a collective emergent phenomenon with a potential for provoking a crisis. Modelling approaches have been deployed since five decades ago; however, the focus was mostly on epidemic behaviour of the rumours which does not take into account the differences between agents. We use social practice theory to model agent decision-making in organizational rumourmongering.

WhatsApp Peer Coaching Lessons for eHealth

WhatsApp was evaluated as a peer coach group support tool in a healthy lifestyle intervention with 15 young professionals. These individuals were time-constrained professionals, so two design challenges were to create enough attractiveness and quality in the peer group interactions. There were three main health domains: food, physical activity, and mental energy. As a result of the 12 week pilot, there were 127 WhatsApp peer coaching inputs.

Automated Configuration of Negotiation Strategies

Bidding and acceptance strategies have a substantial impact on the outcome of negotiations in scenarios with linear additive and nonlinear utility functions. Over the years, it has become clear that there is no single best strategy for all negotiation settings, yet many fixed strategies are still being developed. We envision a shift in the strategy design question from: What is a good strategy?, towards: What could be a good strategy? For this purpose, we developed a method leveraging automated algorithm configuration to find the best strategies for a specific set of negotiation settings.

Improving Confidence in the Estimation of Values and Norms

Autonomous agents (AA) will increasingly be interacting with us in our daily lives. While we want the benefits attached to AAs, it is essential that their behavior is aligned with our values and norms. Hence, an AA will need to estimate the values and norms of the humans it interacts with, which is not a straightforward task when solely observing an agent's behavior. This paper analyses to what extent an AA is able to estimate the values and norms of a simulated human agent (SHA) based on its actions in the ultimatum game.

New Foundations of Ethical Multiagent Systems

Ethics is inherently a multiagent concern. However, research on AI ethics today is dominated by work on individual agents:(1) how an autonomous robot or car may harm or (differentially) benefit people in hypothetical situations (the so-called trolley problems) and (2) how a machine learning algorithm may produce biased decisions or recommendations. The societal framework is largely omitted. To develop new foundations for ethics in AI, we adopt a sociotechnical stance in which agents (as technical entities) help autonomous social entities or principals (people and organizations).

Think Too Fast Nor Too Slow: The Computational Trade-off Between Planning And Reinforcement Learning

Planning and reinforcement learning are two key approaches to sequential decision making. Multi-step approximate real-time dynamic programming, a recently successful algorithm class of which AlphaZero [Silver et al., 2018] is an example, combines both by nesting planning within a learning loop. However, the combination of planning and learning introduces a new question: how should we balance time spend on planning, learning and acting? The importance of this trade-off has not been explicitly studied before.

The Second Type of Uncertainty in Monte Carlo Tree Search

Monte Carlo Tree Search (MCTS) efficiently balances exploration and exploitation in tree search based on count-derived uncertainty. However, these local visit counts ignore a second type of uncertainty induced by the size of the subtree below an action. We first show how, due to the lack of this second uncertainty type, MCTS may completely fail in well-known sparse exploration problems, known from the reinforcement learning community.

Things that help out: designing smart wearables as partners in stress management

We propose an approach to designing smart wearables that act as partners to help people cope with stress in daily life. Our approach contributes to the developing field of smart wearables by addressing how technological capabilities can be designed to establish partnerships that consider the person, the situation, and the appropriate type of support. As such, this study also contributes to healthcare by opening up novel technology-supported routes to stress treatment and care.

A Framework for Reinforcement Learning and Planning

Sequential decision making, commonly formalized as Markov Decision Process optimization, is a key challenge in artificial intelligence. Two successful approaches to MDP optimization are planning and reinforcement learning. Both research fields largely have their own research communities. However, if both research fields solve the same problem, then we should be able to disentangle the common factors in their solution approaches.