Lecture Matthew Botvinick: "Deep meta-learning"

On 18 March, Special Interest Group AI and Behavioral Change, part of the focus area Human-centered Artificial Intelligence, organizes a talk by Prof. Matthew Botvinick titled "Deep meta-learning".

Deep meta-learning

Over the past decade there has been explosive progress in deep learning. However, many see a remaining challenge in 'sample efficiency': deep learning is data hungry, and can be limited in its ability to adapt flexibly and rapidly -- as humans do -- to new task challenges. One approach to this problem is meta-learning or learning-to-learn, where a slow learning process, unfolding over many tasks, equips the learner to learn with increasingly nimbleness. Botvinick will discuss his group's explorations of deep meta-leaning, focusing in particular on the questions of how well meta-learning may scale and of how it may be relevant to human-AI interaction.

Matthew Botvinick

Prof. Botvinick is Director of Neuroscience Research at DeepMind and Honorary Professor of the Gatsby Computational Neuroscience Unit at University College London. He is a leading researcher on the intersection between AI and cognitive neurosciences.