Martin Schmid presents The Story of Applied Reinforcement Learning

On 2025-09-19 11:00:00 at G205, Karlovo náměstí 13, Praha 2
The talk will not be an introduction to reinforcement learning (RL), nor a deep
dive into algorithmic details. After formulating the RL task, I will focus on
how applied reinforcement learning is shaping the world around us, and how these
real-world applications then shape the field itself. My goal is to build
intuition and provide a high-level perspective on the history, current
landscape, and future of reinforcement learning. I hope this will not only be
fun. but will also inspire the audience to consider new applications of
reinforcement learning within their own research domains.

Martin Schmid holds a Ph.D. from algorithmic game theory and he is the founder
of Equilibre Technologies. He previously worked as a project lead and senior
research scientist at DeepMind and at IBM. He is a co-author of "Deepstack:
Expert-level artificial intelligence in heads-up no-limit poker" (Science) and
the "Student of games: A unified learning algorithm for both perfect and
imperfect information games".
Za obsah zodpovídá: Petr Pošík