- Date: 12.03.2019
Location: INRIA Bordeaux Sud-Ouest, Bordeaux
The ideas in this talk emerged from two simple questions. First, what does my RL agent really know vs. what kinds of physics can be known to an organism in a similar environment? Second, what kinds of statistical models might allow an organism to gain such knowledge in a Popperian sense?
- Date: 18.09.2018
Location: University of Sussex, Brighton
The goal of this presentation was to stimulate a discussion on the levels of abstraction, akin to Marr’s levels, for understanding Intrinsic Motivation systems. This line of reasoning is taken from Pierre-Yves Oudeyer’s Computational Theories of Curiosity-driven learning.
- Date: 24.05.2018
Location: École 42, 96 Boulevard Bessières, Paris
This introductory presentation on reinforcement learning was presented to software engineering students at École 42 and followed by a Q-learning exercise focused on RoShamBo.