Working with me?
I am recruiting PhD students and postdocs: see vacancies.
I am happy to supervise master projects, please look here.
I am afraid that I cannot offer internships, sorry!
Current students, please check these guidelines & resources.
My main interests lie in the field of sequential decision making under uncertainty (SDM) and especially ‘the reinforcement learning problem’. This field of SDM studies methods that allow an intelligent system, an agent, to make decisions over time. The reinforcement learning problem (not to confused with a narrow set of techniques like Q-learning, etc.) occurs when an SDM agent does not have a model of its environment, but instead will need to learn this its interaction with it.
Sequential decision making is studied in, and has close ties to many research fields within AI, CS and economics such as: planning, reinforcement learning, optimization, machine learning, regression, classification, graphical models, probabilistic inference and game theory. I also have an interest in multiagent systems, since as soon as we construct intelligent SDM agents, they will need to interact with humans and each other.
Some particular topics that I have done research on:
- Bayesian reinforcement learning under partial observability (in “POMDPs”)
- generative adversarial networks
- reinforcement learning for smart coordination of traffic lights
- decentralized MCTS for teams of robots
- decentralized POMDPs
- exploiting structure in graphical models of interactions
- reasoning about trust in e-commerce settings