Michele Garibbo

Postdoctoral research associate

[w/ Laurence Aitchison (Bristol)]
.

My research interest focuses on applications of (deep) reinforcement learning (RL) methods to neuroscience. In the past, I worked on applications of policy gradients methods to human motor learning as well as purely deep RL projects on value estimation. Lately, my focus has shifted towards trying to understand cognitive planning through the lenses of model-based RL methods. I obtained a master of research in Cognitive Neuroscience at UCL and undertook contract education in Data Science and Knowledge Engineering at the University of Maastricht. There, I also undertook an internship in deep RL. I recently submitted my PhD thesis at the University of Bristol, and I am not a short-term postdoc at the group.