Carlo D’Eramo is a computer science professor at the University of Würzburg who specializes in a form of machine learning called reinforcement learning. This type of machine learning does not require data input from humans, instead the algorithms gain experience by solving tasks in simulation runs and are rewarded for correct decisions. The professor’s group is working on further improving the methods of reinforcement learning, such as teaching the self-learning systems multitasking and transfer learning. Transfer learning is similar to humans transferring knowledge from one activity to another.