Self-Validation: Early Stopping for Single-Instance Deep Generative Priors
Self-Validation: Early Stopping for Single-Instance Deep Generative Priors
复制标题
DOI：
--
复制
发表时间：
2021-10
影响因子：
7
通讯作者：
Taihui Li;Zhong Zhuang;Hengyue Liang;L. Peng;Hengkang Wang;Ju Sun
中科院分区：
文献类型：
--
作者：
Taihui Li;Zhong Zhuang;Hengyue Liang;L. Peng;Hengkang Wang;Ju Sun
最近的工作表明，即使在没有训练数据的情况下，深度生成模型在解决大量图像重建(IR)任务时也具有令人惊讶的有效性。我们将这些模型，如深度图像先验和深度译码，统称为单实例深度生成先验(SIDGP)。然而，成功往往取决于适当的早期停止，到目前为止，这在很大程度上是以临时方式处理的。在本文中，我们利用重建质量的典型钟形趋势，提出了将SIDGP应用于IR时ES的第一个原则性方法。特别是，我们的方法是基于协作训练和自我验证的：原始重建过程由深度自动编码器监控，该编码器使用历史重建图像在线训练，并用于不断验证重建质量。实验表明，在几个IR问题和不同的SIDGP上，我们的自验证方法能够可靠地检测到接近峰值的性能并发出良好的ES点信号。我们的代码可以在https://sun-umn.github.io/Self-Validation/.上找到
Recent works have shown the surprising effectiveness of deep generative models in solving numerous image reconstruction (IR) tasks, even without training data. We call these models, such as deep image prior and deep decoder, collectively as single-instance deep generative priors (SIDGPs). The successes, however, often hinge on appropriate early stopping (ES), which by far has largely been handled in an ad-hoc manner. In this paper, we propose the first principled method for ES when applying SIDGPs to IR, taking advantage of the typical bell trend of the reconstruction quality. In particular, our method is based on collaborative training and self-validation: the primal reconstruction process is monitored by a deep autoencoder, which is trained online with the historic reconstructed images and used to validate the reconstruction quality constantly. Experimentally, on several IR problems and different SIDGPs, our self-validation method is able to reliably detect near-peak performance and signal good ES points. Our code is available at https://sun-umn.github.io/Self-Validation/.