Search Results for author:
Found 9 papers, 3 papers with code
Asymptotic Analysis of Two-Layer Neural Networks after One Gradient Step under Gaussian Mixtures Data with Structure
In this work, we study the training and generalization performance of two-layer neural networks (NNs) after one gradient descent step under structured data modeled by Gaussian mixtures.
Exploring the Precise Dynamics of Single-Layer GAN Models: Leveraging Multi-Feature Discriminators for High-Dimensional Subspace Learning
Subspace learning is a critical endeavor in contemporary machine learning, particularly given the vast dimensions of modern datasets.
Random Features Outperform Linear Models: Effect of Strong Input-Label Correlation in Spiked Covariance Data
Our analysis reveals that a high correlation between inputs and labels is a critical factor enabling the RFM to outperform linear models.
Training Generative Image Super-Resolution Models by Wavelet-Domain Losses Enables Better Control of Artifacts
Although some recent works focused on the differentiation of details and artifacts, this is a very challenging problem and a satisfactory solution is yet to be found.
Trustworthy SR: Resolving Ambiguity in Image Super-resolution via Diffusion Models and Human Feedback
This strategy leverages the high-quality image generation capabilities of DMs, while recognizing the importance of obtaining a single trustworthy solution, especially in use cases, such as identification of specific digits or letters, where generating multiple feasible solutions may not lead to a reliable outcome.
Perception-Distortion Trade-off in the SR Space Spanned by Flow Models
We achieve this by benefiting from a diverse set of feasible photo-realistic solutions in the SR space spanned by flow models.
MMSR: Multiple-Model Learned Image Super-Resolution Benefiting From Class-Specific Image Priors
As a result, the performance of an SR model varies noticeably from image to image over a test set depending on whether characteristics of specific images are similar to those in the training set or not.
Two-stage domain adapted training for better generalization in real-world image restoration and super-resolution
It is well-known that in inverse problems, end-to-end trained networks overfit the degradation model seen in the training set, i. e., they do not generalize to other types of degradations well.
On the Computation of PSNR for a Set of Images or Video
Others compute a single PSNR from the arithmetic mean of frame MSEs for each video.
