Esther Derman – MILA, Canada
Robust Markov decision processes (MDPs) aim to handle changing or partially known system dynamics. To solve them, one typically resorts to robust optimization methods. However, this significantly increases computational complexity and limits scalability in both learning and planning. On the other hand, regularized MDPs show more stability in policy learning without impairing time complexity. Yet, they generally do not encompass uncertainty in the model dynamics. In this talk, I will show how we can learn robust MDPs using proper regularization, so as to reduce planning and learning in robust MDPs to regularized MDPs.
Campus de l'Université de Montréal
2920, chemin de la Tour
Montréal Québec H3T 1J4