Enhancing Reinforcement Learning with discrete interfaces to learn the Dyck Language
This addresses the challenge of learning hierarchical structures with discrete interfaces, which is important for real-world applications but has been difficult for neural networks.
The researchers tackled the problem of training neural networks to use discrete interfaces by enhancing a Reinforcement Learning architecture with discrete interfaces and training it on the generalized Dyck language, resulting in a model that generalizes to sequences an order of magnitude longer than the training data.
Even though most interfaces in the real world are discrete, no efficient way exists to train neural networks to make use of them, yet. We enhance an Interaction Network (a Reinforcement Learning architecture) with discrete interfaces and train it on the generalized Dyck language. This task requires an understanding of hierarchical structures to solve, and has long proven difficult for neural networks. We provide the first solution based on learning to use discrete data structures. We encountered unexpected anomalous behavior during training, and utilized pre-training based on execution traces to overcome them. The resulting model is very small and fast, and generalizes to sequences that are an entire order of magnitude longer than the training data.