Interactive Learning Group
Max Planck Fellow Group
We focus on investigations at the intersection of causal inference and reinforcement learning, in a joint research venture with the Department for Empirical Inference.
The premise behind this is the observation that with the success and explosive adoption of machine learning, we increasingly see systems that do not operate in the standard i.i.d. setting, but interact with and thus intervene in the world. This pertains not only to the virtual world (e.g., recommender systems), but increasingly also to the physical world (internet of things, medical applications, renewable energy systems, etc.). As a consequence, future learning systems must be able to anticipate effects of actions and interventions using observational data. The group is led by Prof. Dr. Andreas Krause; the funding comes from a Max Planck Fellowship.
The groups of Andreas Krause are distributed between ETH Zurich and the Max Planck Institute for Intelligent Systems. Our main webpage is here: Learning & Adaptive Systems Group.