The Neural and machine learning group seeks to understand the principles of learning in the brain. Our work draws inspiration from different disciplines (neuroscience, machine learning, statistics, computer science, physics) across multiple levels of analysis, from single synapse to cross-area interactions via neural circuits. We collaborate with experimental and theoretical/machine learning labs in the UK and internationally.
keywords: Cortical circuits, synaptic plasticity, credit assignment, reinforcement learning, probabilistic inference, machine learning and deep learning.
Laurence Aitchison group uses Bayesian approaches to uncover the theoretical principles behind biological and artificial intelligence, and at the same time we develop novel algorithms for data-analysis and apply them to datasets ranging from calcium-imaging to human behaviour.