QGAN: Quantized Generative Adversarial Networks

Abstract

The intensive computation and memory requirements of generative adversarial neural networks (GANs) hinder its real-world deployment on edge devices such as smartphones. Despite the success in model reduction of CNNs, neural network quantization methods have not yet been studied on GANs, which are mainly faced with the issues of both the effectiveness of quantization algorithms and the instability of training GAN models. In this paper, we start with an extensive study on applying existing successful methods to quantize GANs. Our observation reveals that none of them generates samples with reasonable quality because of the underrepresentation of quantized values in model weights, and the generator and discriminator networks show different sensitivities upon quantization methods. Motivated by these observations, we develop a novel quantization method for GANs based on EM algorithms, named as QGAN. We also propose a multi-precision algorithm to help find the optimal number of bits of quantized GAN models in conjunction with corresponding result qualities. Experiments on CIFAR-10 and CelebA show that QGAN can quantize GANs to even 1-bit or 2-bit representations with results of quality comparable to original models.

References in corpus (12)

• Conditional Generative Adversarial Nets
• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• Large Scale GAN Training for High Fidelity Natural Image Synthesis
• Binarized Neural Networks: Training Deep Neural Networks with Weights and Activations Constrained to +1 or -1
• DoReFa-Net: Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradients
• Progressive Growing of GANs for Improved Quality, Stability, and Variation
• Quantized Neural Networks: Training Neural Networks with Low Precision Weights and Activations
• Mixed Precision Training
• Energy-based Generative Adversarial Network
• Trained Ternary Quantization
• Convolutional Neural Networks using Logarithmic Data Representation
• Scalable Methods for 8-bit Training of Neural Networks

Cited by in corpus (5)

• Training Deep Neural Networks with Joint Quantization and Pruning of Weights and Activations
• GANs Can Play Lottery Tickets Too
• Self-Supervised GAN Compression
• A Survey on GAN Acceleration Using Memory Compression Technique
• Quantized Proximal Averaging Network for Analysis Sparse Coding