# Learning Hierarchical Information Flow with Recurrent Neural Modules

arXiv: https://arxiv.org/abs/1706.05744 · HTML: https://mlsanity.com/paper/1706.05744/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 18 Jun 2017.

## Built on by (most cited first, 1 of 1)

- [Oscillatory Fourier Neural Network: A Compact and Efficient Architecture for Sequential Processing](https://mlsanity.com/paper/2109.13090/) (2021)

## Abstract (arXiv)

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.