Spectral Mixture

Spectral mixture models are used to represent complex signals or data by decomposing them into a combination of simpler spectral components. Current research focuses on improving the efficiency and scalability of these models, particularly within Gaussian process frameworks, for applications like solving partial differential equations and graph clustering. This involves developing novel kernel designs, such as spectral mixture kernels and their variations, and employing techniques like sparsity-aware optimization and structured automatic differentiation to handle high-dimensional data. These advancements enhance the accuracy and applicability of spectral mixture methods across diverse fields, including machine learning, signal processing, and computer vision.

Papers

November 8, 2023

Solving High Frequency and Multi-Scale PDEs with Gaussian Processes
Shikai Fang, Madison Cooley, Da Long, Shibo Li, Robert Kirby, Shandian Zhe
Gaussian Process High Frequency Gaussian Mixture Spectral Mixture

November 3, 2023

Spectral Clustering of Attributed Multi-relational Graphs
Ylli Sadikaj, Yllka Velaj, Sahar Behzadi, Claudia Plant
Graph Structure Spectral Clustering Joint Embedding Multiple Authorship Multi Relational Spectral Mixture

November 2, 2023

Scattering Vision Transformer: Spectral Mixing Matters
Badri N. Patro, Vijay Srinivas Agneeswaran
Vision Transformer Supervised ImageNet Vision Task Scattering Network Spectral Mixture

September 15, 2023

Sparsity-Aware Distributed Learning for Gaussian Processes with Linear Multiple Kernel
Richard Cornelius Suwandi, Zhidi Lin, Feng Yin, Zhiguo Wang, Sergios Theodoridis
LeArning Abstract Gaussian Process Hyperparameter Optimization Kernel Learning Sparsity Aware Linear Kernel Spectral Mixture

September 22, 2022

AcroFOD: An Adaptive Method for Cross-domain Few-shot Object Detection
Yipeng Gao, Lingxiao Yang, Yunmu Huang, Song Xie, Shiyong Li, Wei-shi Zheng
Data Augmentation Object Detector Adaptive Method Feature Distribution Cross Domain Few Shot Spectral Mixture

June 16, 2022

Scalable First-Order Bayesian Optimization via Structured Automatic Differentiation
Sebastian Ament, Carla Gomes
Bayesian Optimization Automatic Differentiation Kernel Matrix Continuous Kernel Spectral Mixture First Order Bayesian Optimization

February 18, 2022

Nonstationary multi-output Gaussian processes via harmonizable spectral mixtures
Matías Altamirano, Felipe Tobar
Gaussian Mixture Multi Output Gaussian Process Stationary Kernel Spectral Mixture