SpecNet: Spectral Domain Convolutional Neural Network
arXiv:1905.10915
Abstract
The memory consumption of most Convolutional Neural Network (CNN) architectures grows rapidly with increasing depth of the network, which is a major constraint for efficient network training on modern GPUs with limited memory, embedded systems, and mobile devices. Several studies show that the feature maps (as generated after the convolutional layers) are the main bottleneck in this memory problem. Often, these feature maps mimic natural photographs in the sense that their energy is concentrated in the spectral domain. Although embedding CNN architectures in the spectral domain is widely exploited to accelerate the training process, we demonstrate that it is also possible to use the spectral domain to reduce the memory footprint, a method we call Spectral Domain Convolutional Neural Network (SpecNet) that performs both the convolution and the activation operations in the spectral domain. The performance of SpecNet is evaluated on three competitive object recognition benchmark tasks (CIFAR-10, SVHN, and ImageNet), and compared with several state-of-the-art implementations. Overall, SpecNet is able to reduce memory consumption by about 60% without significant loss of performance for all tested networks.
arXiv commentAccepted by ICASSP 21. Contact author: Bochen Guan ([email protected])
References in corpus (11)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Exploiting Linear Structure Within Convolutional Networks for Efficient Evaluation
- One weird trick for parallelizing convolutional neural networks
- A Survey of Model Compression and Acceleration for Deep Neural Networks
- Training Deep Nets with Sublinear Memory Cost
- Diffusion-Convolutional Neural Networks
- Fast Training of Convolutional Networks through FFTs
- Wavelet Convolutional Neural Networks for Texture Classification
- The Power of Sparsity in Convolutional Neural Networks
- Memory-Efficient Implementation of DenseNets
- Efficient Memory Management for GPU-based Deep Learning Systems