Prune the Convolutional Neural Networks with Sparse Shrink
This addresses the computational and memory burdens of CNNs for embedded applications, representing an incremental improvement in model compression.
The paper tackles the problem of deploying Convolutional Neural Networks (CNNs) on embedded devices by proposing a 'Sparse Shrink' algorithm to prune redundant feature maps, resulting in a 56.77% reduction in parameters and 73.84% reduction in multiplications with only minor accuracy loss on CIFAR-100.
Nowadays, it is still difficult to adapt Convolutional Neural Network (CNN) based models for deployment on embedded devices. The heavy computation and large memory footprint of CNN models become the main burden in real application. In this paper, we propose a "Sparse Shrink" algorithm to prune an existing CNN model. By analyzing the importance of each channel via sparse reconstruction, the algorithm is able to prune redundant feature maps accordingly. The resulting pruned model thus directly saves computational resource. We have evaluated our algorithm on CIFAR-100. As shown in our experiments, we can reduce 56.77% parameters and 73.84% multiplication in total with only minor decrease in accuracy. These results have demonstrated the effectiveness of our "Sparse Shrink" algorithm.