Jean-Paul Noel

This group records from tens of thousands of neurons in awake and behaving rodents. The researchers then try to relate the neural activity they observe with behavior. The main question of interest is: how do we map sensory observations to hidden objects in the environment? Essentially, they are determining how to hypothesize latent causes to the observations. This is closely related to the credit assignment problem in AI. The group uses supercomputers to fit models to both behavior and neural activity.