计算机科学
透视图(图形)
发电机(电路理论)
样品(材料)
人工智能
培训(气象学)
选择(遗传算法)
机器学习
训练集
自适应系统
班级(哲学)
数据挖掘
信号发生器
作者
Zhaoyu Zhang,Hua Yang,Guanxiong Sun,Hui Wang,Seán McLoone
标识
DOI:10.1109/tmm.2026.3703344
摘要
Recently, many studies have highlighted that training data-efficient generative adversarial networks (DE-GANs) suffers from the overfitting of the discriminator ($D$) problem. However, how this issue theoretically influences the training of the generator ($G$) remains unclear. In this paper, we unveil a novel insight in DE-GANs training, regarding the significance of useful gradients of$G$. Notably, the useful gradients of$G$are those computed on generated samples closer to the real data distribution, which helps the generator to better learn the real data distribution. As the overfitting degree of$D$increases, the gradients of the$G$become progressively less useful, which hinders DE-GANs training. Based on this novel insight, we propose adaptive Top-k (ATop-k), a novel adaptive sample selection strategy for DE-GANs to provide more useful gradients for$G$by selecting fake samples to update$G$during training. Concretely, only the Top-k high-score fake samples are used to update$G$, where the value of k is controlled by the overfitting degree of$D$adaptively. Extensive experiments on several datasets demonstrate that ATop-k can effectively achieve better performance across different DE-GANs and deliver state-of-the-art performance compared with other GANs and diffusion models.
科研通智能强力驱动
Strongly Powered by AbleSci AI