Hierarchical Multiscale Recurrent Neural Networks
This work addresses a fundamental problem in sequence modeling for AI researchers, though it appears incremental as it builds on existing multiscale approaches without claiming major breakthroughs.
The paper tackles the challenge of learning hierarchical and temporal representations in recurrent neural networks by proposing hierarchical multiscale RNNs, which capture latent hierarchical structures in sequences using a novel update mechanism, and evaluates the model on character-level language and handwriting sequence modeling tasks.
Learning both hierarchical and temporal representation has been among the long-standing challenges of recurrent neural networks. Multiscale recurrent neural networks have been considered as a promising approach to resolve this issue, yet there has been a lack of empirical evidence showing that this type of models can actually capture the temporal dependencies by discovering the latent hierarchical structure of the sequence. In this paper, we propose a novel multiscale approach, called the hierarchical multiscale recurrent neural networks, which can capture the latent hierarchical structure in the sequence by encoding the temporal dependencies with different timescales using a novel update mechanism. We show some evidence that our proposed multiscale architecture can discover underlying hierarchical structure in the sequences without using explicit boundary information. We evaluate our proposed model on character-level language modelling and handwriting sequence modelling.