The Uncanny Similarity of Recurrence and Depth

Avi Schwarzschild, Arjun Gupta, Amin Ghiasi, Micah Goldblum, Tom Goldstein

Keywords: Deep learning, recurrent networks, depth

Abstract: It is widely believed that deep neural networks contain layer specialization, wherein networks extract hierarchical features representing edges and patterns in shallow layers and complete objects in deeper layers. Unlike common feed-forward models that have distinct filters at each layer, recurrent networks reuse the same parameters at various depths. In this work, we observe that recurrent models exhibit the same hierarchical behaviors and the same performance benefits as depth despite reusing the same filters at every recurrence. By training models of various feed-forward and recurrent architectures on several datasets for image classification as well as maze solving, we show that recurrent networks have the ability to closely emulate the behavior of non-recurrent deep models, often doing so with far fewer parameters.

One-sentence Summary: We show quantitatively and qualitatively that recurrent models have the same behaviors as feed-forward networks despite reusing parameters at each recurrence.