arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2104.09223cs.CV
收藏

从粗到精搜索高效生成对抗网络

Coarse-to-Fine Searching for Efficient Generative Adversarial Networks

Tsinghua University(清华大学);Noah’s Ark Lab, Huawei Technologies(华为技术有限公司诺亚方舟实验室)

Jiahao Wang, Han Shu, Weihao Xia, Yujiu Yang, Yunhe Wang

摘要:

本文研究了开发高效生成器网络的神经结构搜索(NAS)问题。与用于视觉识别任务的深度模型相比,生成对抗网络(GAN)通常被设计用于进行各种复杂图像生成。我们首先发现一个完整的生成器网络的搜索空间,包括三个维度,即,充分挖掘网络性能的路径、运营商、渠道。为了降低搜索成本,我们探索了一种由粗到精的搜索策略,将整个搜索过程分为三个子优化问题。此外,一个公平的超网训练方法,以确保所有的子网络可以公平和稳定的更新。基准测试的实验结果表明,我们可以提供更好的图像质量和更低的计算成本比国家的最先进的方法的生成器网络。例如,使用我们的方法,在整个edges-to-shoes数据集上只需大约8个GPU小时即可获得2.56 MB的模型,FID得分为24.13,在整个Urban 100数据集上只需10个GPU小时即可获得1.49 MB的模型,PSNR得分为24.94。

英文摘要:

This paper studies the neural architecture search (NAS) problem for developing efficient generator networks. Compared with deep models for visual recognition tasks, generative adversarial network (GAN) are usually designed to conduct various complex image generation. We first discover an intact search space of generator networks including three dimensionalities, i.e., path, operator, channel for fully excavating the network performance. To reduce the huge search cost, we explore a coarse-to-fine search strategy which divides the overall search process into three sub-optimization problems accordingly. In addition, a fair supernet training approach is utilized to ensure that all sub-networks can be updated fairly and stably. Experiments results on benchmarks show that we can provide generator networks with better image quality and lower computational costs over the state-of-the-art methods. For example, with our method, it takes only about 8 GPU hours on the entire edges-to-shoes dataset to get a 2.56 MB model with a 24.13 FID score and 10 GPU hours on the entire Urban100 dataset to get a 1.49 MB model with a 24.94 PSNR score.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑