Embedded Files
 Explainable AI for Sequential Decision Making, or How to Leave the Car at Home
How can AI solutions have an impact on sustainability goals, such as reducing car traffic? The actual obstacle is often not the capability of AI systems to effectively solve optimization, scheduling, and planning tasks. Instead, the problem is the AI's inability to explain, discuss and refine their solutions together with human users, their inability to inspire trust - and actually get used. In this talk, I will first talk about individual urban mobility, and show how more flexible and personalized XAI could help increase trust in routing solutions and make the city more accessible. Then, I will talk about similar problems at larger scale, and present an XAI approach for a paratransit planning system whose operators need to schedule and batch trip requests across a vehicle fleet.
Eindhoven University of Technology and Centrum Wiskunde & Informatica, Netherlands
Hendrik Baier is Assistant Professor at Eindhoven University of Technology (TU/e). He is interested in explainable sequential decision making, in order to create AI agents that can collaborate with human users and solve their complex real-world problems. To achieve this, his research focuses on planning, required for acting towards long-term goals; on learning, required for acting in unknown environments; and on the explainability of planning and learning, required for successful human-AI interaction. He has leading roles in national and international collaborative projects with industry partners in sectors such as logistics, transportation, and sustainable energy.
At IJCAI-ECAI 2026, Hendrik is organizing the long-running Workshop on Explainable Artificial Intelligence (XAI), which is relevant for anyone wondering how to make their AI systems more understandable, accessible, and trustworthy.
Page updated
Google Sites
Report abuse