Description
Stable Rank Normalization (SRN) is a weight-normalization scheme which minimizes the stable rank of a linear operator. It simultaneously controls the Lipschitz constant and the stable rank of a linear operator. Stable rank is a softer version of the rank operator and is defined as the squared ratio of the Frobenius norm to the spectral norm.
Papers Using This Method
Exploring Semantic Feature Discrimination for Perceptual Image Super-Resolution and Opinion-Unaware No-Reference Image Quality Assessment2025-03-25GUMBEL-NERF: Representing Unseen Objects as Part-Compositional Neural Radiance Fields2024-10-27Regularized Multi-Decoder Ensemble for an Error-Aware Scene Representation Network2024-07-26PICASSO: A Feed-Forward Framework for Parametric Inference of CAD Sketches via Rendering Self-Supervision2024-07-18LeanGaussian: Breaking Pixel or Point Cloud Correspondence in Modeling 3D Gaussians2024-04-25Adaptively Placed Multi-Grid Scene Representation Networks for Large-Scale Data Visualization2023-07-16Towards a general purpose machine translation system for Sranantongo2022-12-13Line Drawing Guided Progressive Inpainting of Mural Damage2022-11-12Novel View Synthesis with Diffusion Models2022-10-06An Efficient Two-Stream Network for Isolated Sign Language Recognition Using Accumulative Video Motion2022-09-06A Hierarchical Speaker Representation Framework for One-shot Singing Voice Conversion2022-06-28Reasoning Structural Relation for Occlusion-Robust Facial Landmark Localization2021-12-19A NEW BACKBONE FOR HYPERSPECTRAL IMAGE RECONSTRUCTION2021-09-29CodeNeRF: Disentangled Neural Radiance Fields for Object Categories2021-09-03A Simple and Efficient Reconstruction Backbone for Snapshot Compressive Imaging2021-08-17Sequential Random Network for Fine-grained Image Classification2021-03-12Stable Rank Normalization for Improved Generalization in Neural Networks and GANs2019-06-11