Minimal Effort Back Propagation for Convolutional Neural Networks
This reduces computational resources for training CNNs, but it is incremental as it adapts an existing method to a new architecture.
The paper tackles the high computational cost of back propagation in convolutional neural networks by extending a gradient sparsification technique to CNNs, achieving comparable or better performance while using only 5% of the gradients.
As traditional neural network consumes a significant amount of computing resources during back propagation, \citet{Sun2017mePropSB} propose a simple yet effective technique to alleviate this problem. In this technique, only a small subset of the full gradients are computed to update the model parameters. In this paper we extend this technique into the Convolutional Neural Network(CNN) to reduce calculation in back propagation, and the surprising results verify its validity in CNN: only 5\% of the gradients are passed back but the model still achieves the same effect as the traditional CNN, or even better. We also show that the top-$k$ selection of gradients leads to a sparse calculation in back propagation, which may bring significant computational benefits for high computational complexity of convolution operation in CNN.