Bridging the Gaps Between Residual Learning, Recurrent Neural Networks and Visual Cortex

Code

• github.com/MindSpore-scientific-2/code-10/tree/main/shallow-rnns
  mindspore★ 0
• github.com/MindSpore-scientific/code-13/tree/main/shallow-rnns
  mindspore★ 0
• github.com/MindSpore-scientific-2/code-11/tree/main/shallow-rnns
  mindspore★ 0
• github.com/ry/tensorflow-resnet
  tf★ 0

Abstract

We discuss relations between Residual Networks (ResNet), Recurrent Neural Networks (RNNs) and the primate visual cortex. We begin with the observation that a special type of shallow RNN is exactly equivalent to a very deep ResNet with weight sharing among the layers. A direct implementation of such a RNN, although having orders of magnitude fewer parameters, leads to a performance similar to the corresponding ResNet. We propose 1) a generalization of both RNN and ResNet architectures and 2) the conjecture that a class of moderately deep RNNs is a biologically-plausible model of the ventral stream in visual cortex. We demonstrate the effectiveness of the architectures by testing them on the CIFAR-10 and ImageNet dataset.

Reproductions