In this Talk, Matteo Marsili will discuss how abstract representations emerge in Deep Belief Networks (DBN) trained on benchmark datasets. As the data is processed by deeper and deeper layers, features are detected and removed, transferring more and more “context-invariant” information to deeper layers. They show that the representation approaches a universal model determined by the principle of maximal relevance. Relevance quantifies the uncertainty on the model of the data, thus suggesting that “meaning” – i.e. syntactic information – is that part of the data which is not yet captured by a model. Their analysis shows that shallow layers are well described by models with pairwise interactions, which provide a representation of the data in terms of generic, low-order features. They also show that plasticity increases with depth, in a similar way as it does in the brain.