Reinforcement Learning: Bringing Together Computation, Behavior and Neural Coding

  • Anne Collins | Helen Wills Neuroscience Institute at UC Berkeley
Reinforcement learning carries subtly different meanings in machine learning, cognitive science and neuroscience. In this talk, I will try to clarify in which ways the concepts overlap and in which ways they differ. I will present examples from my research of how machine learning RL advances our understanding of brain and behavior, but also of how exploring the mechanisms our brains provide for learning may also inspire better machine learning algorithms.
See presentation slides here: https://www.microsoft.com/en-us/research/wp-content/uploads/2018/11/Bringing-Together-Computation-Behavior-and-Neural-Coding-SLIDES.pdf
Speaker Details

Anne Collins is an Assistant Professor in the department of Psychology and a member of the Helen Wills Neuroscience Institute at UC Berkeley. Her research interests include Human learning, decision-making and executive functions; Computational modeling at multiple levels (cognitive and neuroscience); Behavioral, EEG, drug and genes studies in healthy or patient populations. She studies the multiple neuro-cognitive processes that contribute to learning, decision-making and executive functions.

Her lab uses quantitative modeling at different levels of description (e.g. reinforcement learning, Bayesian inference, and neural network modeling) to study the computations performed by cognitive or neural systems. To inform the quantitative theories developed, the lab performs behavioral, as well as EEG, drug or genes experiments in healthy or patient adult populations.

Watch Next