EVENTS

We are excited to welcome Chris Summerfield from U Oxford to Hamburg for at talk next week. Chris is a fantastic speaker and the talk is open to everyone interested. 

 

I will discuss the computational mechanisms that underlie learning in biological and artificial networks. I will argue that both humans and neural networks trained with gradient descent show simplicity biases, a tendency to learn generalities before specifics in a dataset. I will show that this path-dependent learning explains why humans benefit from curricula that prioritise simplicity before complexity. I will discuss data showing that the course of learning and development is supported by a transition from high-dimensional representations, permitting flexible computation, to low-dimensional representations,  permitting generalisation and rule-learning. Finally,  I will provide evidence that curricula help ‘unlearn’ ingrained patterns and undo established learning, to keep learning flexible over the lifespan.

 

Institutions

• Nicolas Schuck's Lab