Skip to yearly menu bar Skip to main content


Poster

Learning Hierarchical Information Flow with Recurrent Neural Modules

Danijar Hafner · Alexander Irpan · James Davidson · Nicolas Heess

Pacific Ballroom #105

Keywords: [ Model Selection and Structure Learning ] [ Recurrent Networks ] [ Memory-Augmented Neural Networks ]


Abstract:

We propose ThalNet, a deep learning model inspired by neocortical communication via the thalamus. Our model consists of recurrent neural modules that send features through a routing center, endowing the modules with the flexibility to share features over multiple time steps. We show that our model learns to route information hierarchically, processing input data by a chain of modules. We observe common architectures, such as feed forward neural networks and skip connections, emerging as special cases of our architecture, while novel connectivity patterns are learned for the text8 compression task. Our model outperforms standard recurrent neural networks on several sequential benchmarks.

Live content is unavailable. Log in and register to view live content