Please note that this newsitem has been archived, and may contain outdated information or links.
16 February 2024, Formalisation, Optimisation, Algorithms, Mechanisms (FOAM), Matthijs Spaan
Speaker: Matthijs Spaan
Title: Exploiting Epistemic Uncertainty for Deep Exploration in Reinforcement Learning
Date: Friday 16 February 2024
Time: 15:00-16:25
Location: Room L3.33, ILLC Lab42, Science Park 900, Amsterdam
In this talk I discuss how estimating and propagating epistemic uncertainty benefits generalization and deep exploration in reinforcement learning (RL) by focusing on two recent contributions. First, I consider model-free distributional RL, which aims to learn the distribution of returns rather than their expected value. Second, I discuss how propagating epistemic uncertainty estimates can be leveraged in a model-based RL setting, by embedding them in Monte-Carlo Tree Search (MCTS).
For more information, see
https://events.illc.uva.nl/FOAM/posts/talk11/
or contact Gregor Behnke at g.behnke at uva.nl, or Ronald de Haan at r.dehaan at uva.nl.
Please note that this newsitem has been archived, and may contain outdated information or links.