Deep Fried Convnets
This addresses memory constraints for deploying deep neural networks in GPUs or embedded devices, representing an incremental improvement in parameter efficiency.
The paper tackles the problem of high memory consumption in fully connected layers of deep convolutional neural networks by replacing them with a single Fastfood layer, achieving substantial memory reduction on MNIST and ImageNet with no drop in predictive performance.
The fully connected layers of a deep convolutional neural network typically contain over 90% of the network parameters, and consume the majority of the memory required to store the network parameters. Reducing the number of parameters while preserving essentially the same predictive performance is critically important for operating deep neural networks in memory constrained environments such as GPUs or embedded devices. In this paper we show how kernel methods, in particular a single Fastfood layer, can be used to replace all fully connected layers in a deep convolutional neural network. This novel Fastfood layer is also end-to-end trainable in conjunction with convolutional layers, allowing us to combine them into a new architecture, named deep fried convolutional networks, which substantially reduces the memory footprint of convolutional networks trained on MNIST and ImageNet with no drop in predictive performance.