Theoretical Neuroscience Group
The brain is the most complex known object and is able to perform remarkable computations such as tracking the position of a moving prey in a rich environment or producing language. However, despite the intense efforts invested towards its understanding (there are more than 100'000 papers published every year in neuroscience), it remains unclear what are the fundamental computational principles implemented by the brain. We are interested in discovering those principles at three different levels: the single synapse level, the neural network level and the behavioural level. Since noise plays a key role at those different levels, we express those computational principles from a Bayesian perspective and compare the implementation predictions of our models with experimental data.