Score-Guided Intermediate Layer Optimization: Fast Langevin Mixing for Inverse Problems
Score-Guided Intermediate Layer Optimization: Fast Langevin Mixing for Inverse Problems
复制标题
DOI：
10.48550/arxiv.2206.09104
复制
发表时间：
2022-06
期刊：
影响因子：
--
通讯作者：
Giannis Daras;Y. Dagan;A. Dimakis;C. Daskalakis
中科院分区：
文献类型：
--
作者：
Giannis Daras;Y. Dagan;A. Dimakis;C. Daskalakis
我们证明了快速混合，并描述了Langevin算法的平稳分布，用于反转随机加权DNN生成器。这一结果将Hand和Voroninski的工作从有效的反演扩展到有效的后验采样。在实践中，为了提高表现力，我们建议在预训练的生成模型的潜在空间中进行后验采样。为了实现这一目标，我们在StyleGAN-2的潜在空间中训练了一个基于分数的模型，并使用它来解决逆问题。我们的框架，分数引导的中间层优化（SGILO），通过在中间层中用生成先验代替稀疏正则化来扩展先前的工作。实验上，我们获得了显着的改进，在以前的国家的最先进的，特别是在低测量制度。
We prove fast mixing and characterize the stationary distribution of the Langevin Algorithm for inverting random weighted DNN generators. This result extends the work of Hand and Voroninski from efficient inversion to efficient posterior sampling. In practice, to allow for increased expressivity, we propose to do posterior sampling in the latent space of a pre-trained generative model. To achieve that, we train a score-based model in the latent space of a StyleGAN-2 and we use it to solve inverse problems. Our framework, Score-Guided Intermediate Layer Optimization (SGILO), extends prior work by replacing the sparsity regularization with a generative prior in the intermediate layer. Experimentally, we obtain significant improvements over the previous state-of-the-art, especially in the low measurement regime.