Seminar by Vedran Dunjko (LIACS, University of Leiden) on Reinforcement learning in the quantum domain.

Reinforcement learning (RL) is an aspect of machine learning often
utilized in the context of robotics and artificial intelligence. In
comparison to other canonical machine learning modes, namely supervised
and unsupervised learning, RL has received comparatively little
attention from the quantum information processing (QIP) community.
In this review-style talk, I will discuss this interactive mode of
learning, and briefly present its basics, bottlenecks and perspectives.
Following this, I will discuss the potential of the interplay between RL
and QIP, and present a few results discussing how RL methodology can help in
quantum information settings, and, conversely, how QIP methods may
provide enhancements for this increasingly relevant learning model.