An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling
This challenges the common association between sequence modeling and recurrent networks, suggesting convolutional networks as a better starting point for practitioners, though it is incremental in comparing existing methods.
The paper tackles the problem of selecting architectures for sequence modeling by systematically evaluating generic convolutional and recurrent networks across standard tasks, finding that a simple convolutional architecture outperforms LSTMs in diverse datasets and demonstrates longer effective memory.
For most deep learning practitioners, sequence modeling is synonymous with recurrent networks. Yet recent results indicate that convolutional architectures can outperform recurrent networks on tasks such as audio synthesis and machine translation. Given a new sequence modeling task or dataset, which architecture should one use? We conduct a systematic evaluation of generic convolutional and recurrent architectures for sequence modeling. The models are evaluated across a broad range of standard tasks that are commonly used to benchmark recurrent networks. Our results indicate that a simple convolutional architecture outperforms canonical recurrent networks such as LSTMs across a diverse range of tasks and datasets, while demonstrating longer effective memory. We conclude that the common association between sequence modeling and recurrent networks should be reconsidered, and convolutional networks should be regarded as a natural starting point for sequence modeling tasks. To assist related work, we have made code available at http://github.com/locuslab/TCN .