Negative Distance Kernel

Negative distance kernels are being explored to improve various machine learning tasks by leveraging the information contained in the distances between data points, rather than solely focusing on their similarities. Current research emphasizes applications in areas such as data imputation, where negative distance kernels enhance the accuracy and efficiency of diffusion models, and in clustering and goodness-of-fit testing, particularly for complex data structures like spherical data. These advancements are improving the performance of algorithms in diverse fields, including image generation, scene text detection, and manifold learning, by providing more robust and efficient methods for handling high-dimensional data and complex geometries.

Papers

June 22, 2024

Rethinking the Diffusion Models for Numerical Tabular Data Imputation from the Perspective of Wasserstein Gradient Flow
Zhichao Chen, Haoxuan Li, Fangyikang Wang, Odin Zhang, Hu Xu, Xiaoyu Jiang, Zhihuan Song, Eric H. Wang
Diffusion Model Visual Perspective Wasserstein Gradient Flow Deep Imputation Tabular Data Imputation Negative Distance Kernel

February 3, 2024

Goodness-of-Fit and Clustering of Spherical Data: the QuadratiK package in R and Python
Giovanni Saraceno, Marianthi Markatou, Raktim Mukhopadhyay, Mojgan Golzy
Clustering Result Goodness of Fit Spherical Data Negative Distance Kernel

January 22, 2024

EK-Net:Real-time Scene Text Detection with Expand Kernel Distance
Boyuan Zhu, Fagui Liu, Xi Chen, Quan Tang
Scene Text Detection Instance Detection Kernel Distance Negative Distance Kernel

October 4, 2023

Posterior Sampling Based on Gradient Flows of the MMD with Negative Distance Kernel
Paul Hagemann, Johannes Hertrich, Fabian Altekrüger, Robert Beinert, Jannis Chemseddine, Gabriele Steidl
Gradient Flow Conditional Generative Posterior Sampling Maximum Mean Discrepancy Wasserstein Gradient Flow Negative Distance Kernel

October 31, 2022

Study of Manifold Geometry using Multiscale Non-Negative Kernel Graphs
Carlos Hurtado, Sarath Shekkizhar, Javier Ruiz-Hidalgo, Antonio Ortega
Study Feature Class Manifold High Dimensional Space Data Manifold Graph Regression Geometric Estimation Kernel Graph Negative Distance Kernel

Negative Distance Kernel

Papers

Rethinking the Diffusion Models for Numerical Tabular Data Imputation from the Perspective of Wasserstein Gradient Flow

Goodness-of-Fit and Clustering of Spherical Data: the QuadratiK package in R and Python

EK-Net:Real-time Scene Text Detection with Expand Kernel Distance

Posterior Sampling Based on Gradient Flows of the MMD with Negative Distance Kernel

Study of Manifold Geometry using Multiscale Non-Negative Kernel Graphs