Jakub Grudzien Kuba

Kuba is a PhD student at Berkeley AI Research, advised by Pieter Abbeel. His goal is to discover the underlying algorithmic principles of intelligence and use them to develop robust AI systems that learn to solve problems from variously acquired data. His interests include Reinforcement Learning, Inverse Reinforcement Learning, and Meta-Learning.

Before coming to Berkeley, Kuba completed his BSc in Mathematics with Mathematical Computation at Imperial College London and MSc in Statistical Science at University of Oxford. In London, Kuba worked with Yaodong Yang and developed several theorems and algorithms for Multi-Agent RL. In Oxford, he worked with Jakob Foerster where he developed Mirror Learning – a theory of policy optimization for RL.

For more, see his Google Scholar.