LGCVMLFeb 28, 2023

Balanced Training for Sparse GANs

arXiv:2302.14670v214 citationsh-index: 54
Originality Incremental advance
AI Analysis

This work addresses computational efficiency in GAN training, which is an incremental improvement for researchers and practitioners in generative modeling.

The paper tackles the challenge of applying dynamic sparse training to GANs by proposing a balance ratio metric and a method called ADAPT to control it, achieving a good trade-off between performance and computational cost on multiple datasets.

Over the past few years, there has been growing interest in developing larger and deeper neural networks, including deep generative models like generative adversarial networks (GANs). However, GANs typically come with high computational complexity, leading researchers to explore methods for reducing the training and inference costs. One such approach gaining popularity in supervised learning is dynamic sparse training (DST), which maintains good performance while enjoying excellent training efficiency. Despite its potential benefits, applying DST to GANs presents challenges due to the adversarial nature of the training process. In this paper, we propose a novel metric called the balance ratio (BR) to study the balance between the sparse generator and discriminator. We also introduce a new method called balanced dynamic sparse training (ADAPT), which seeks to control the BR during GAN training to achieve a good trade-off between performance and computational cost. Our proposed method shows promising results on multiple datasets, demonstrating its effectiveness.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes