Résumé |
My talk has two aims: Firstly, being a very rapid introduction to computational neuroscience for physicists,
with a focus on how to go from a single neurone to recurrent neural networks as a way to model neuronal
activity. And secondly, to explain more specifically the research questions my current group with Srdjan
Ostojic is interested in, which are centred around modelling cognitive tasks with low-rank recurrent neural
networks. In this context, I will also mention my own ongoing work on a specific cognitive task, a so-called
context dependent decision making task. I will end with some reflection how one has to adapt ones habits
when going from physics to neuroscience/biology, and where I think its good to stay a physicist. |