Towards Adaptive and Stable Compositional Assemblies of Recurrent Neural Network Modules

Recurrent neural networks (RNNs) are computational models regarded as dynamical systems.Modularity is a key ingredient of complex systems.Thus, the composition of RNN modules provides a simple paradigm for building complex computational models, with the potential to approach the human brain capability.We devise strategies for training RNNs assembled into a larger RNN of RNNs, provided with theoretical guarantees of stability that hold during training for the composed global network.Experiments on pixel-by-pixel image classification benchmarks prove the effectiveness of this approach.