Quantization of Generative Adversarial Networks for Efficient Inference: a Methodological Study

TL;DR

This paper investigates the application of quantization techniques to GANs, demonstrating successful 4/8-bit weight and 8-bit activation quantization across multiple architectures without quality loss, enabling efficient inference on edge devices.

Contribution

It provides the first extensive experimental analysis of quantization methods on GANs, offering practical recipes for effective low-bit inference.

Findings

Quantization can be successfully applied to GANs with minimal quality loss.

Practical quantization recipes for StyleGAN, SAGAN, and CycleGAN are proposed.

GANs can be efficiently deployed on edge devices using 4/8-bit weights and 8-bit activations.

Abstract

Generative adversarial networks (GANs) have an enormous potential impact on digital content creation, e.g., photo-realistic digital avatars, semantic content editing, and quality enhancement of speech and images. However, the performance of modern GANs comes together with massive amounts of computations performed during the inference and high energy consumption. That complicates, or even makes impossible, their deployment on edge devices. The problem can be reduced with quantization -- a neural network compression technique that facilitates hardware-friendly inference by replacing floating-point computations with low-bit integer ones. While quantization is well established for discriminative models, the performance of modern quantization techniques in application to GANs remains unclear. GANs generate content of a more complex structure than discriminative models, and thus quantization of GANs is significantly more challenging. To tackle this problem, we perform an extensive experimental study of state-of-art quantization techniques on three diverse GAN architectures, namely StyleGAN, Self-Attention GAN, and CycleGAN. As a result, we discovered practical recipes that allowed us to successfully quantize these models for inference with 4/8-bit weights and 8-bit activations while preserving the quality of the original full-precision models.