AI for decision-making
The fast development of AI technologies presents numerous opportunities for industry and society. At the same time, the need for trustworthy AI systems is at an all-time high. Within the AI for Decision Making (AI4DM) group, we develop鈥痠nnovative AI鈥痬ethods and鈥痶ools鈥痶hat鈥痚mpower users to tackle complex decision-making challenges in industry and society,鈥痑iming to not only provide more efficient and better solutions, but also doing so in a transparent and鈥痳esponsible鈥痺ay.
More specifically, the鈥疉I4DM鈥痜ocuses on鈥痶he following research topics:
- Advanced AI Techniques for Data-driven Optimization: We focus on the development and application of machine learning (ML), deep learning (DL), reinforcement learning (RL), optimization, planning, and search algorithms to solve complex decision-making problems in industry and society.
- Explainable AI (XAI) and Fairness in Decision Support Systems: While pushing the state of the art in AI-supported decision-making, we emphasize transparency, contestability, and fairness of both machine learning models and optimization algorithms.
- Generative AI (GAI) for Enhanced Predictive and Decision-Making Models: We investigate the use of generative AI to improve prediction models and support sequential decision-making and optimization processes across various domains, for example by making use of the world knowledge encoded in Large Language Models.
- Efficient Algorithm Adaptation in AI decision-making models: We create tools and methodologies for efficient parameter tuning in AI decision-making approaches, ensuring that they can robustly adapt to different contexts and applications, while maintaining effectiveness.
- Sustainable AI techniques supporting green transformation: We develop energy-efficient and computationally efficient models for AI decision-making and optimization, for example using surrogate modelling, in order to reduce their environmental impact and achieve long-term sustainability.