Linear Kernel

The linear kernel is a fundamental component in machine learning, particularly within kernel methods like Support Vector Machines and Gaussian Processes, aiming to efficiently capture linear relationships within data. Current research explores its application in diverse contexts, including text classification, 3D perception (LiDAR data), and protein/molecule analysis, often in conjunction with other kernels or within hybrid models that combine linear and non-linear approaches. This focus stems from the linear kernel's computational efficiency and interpretability, making it valuable for large-scale datasets and applications requiring explainable AI, while ongoing work addresses limitations by integrating it into more sophisticated architectures or improving its performance through techniques like conformal transformations.

Papers

August 2, 2024

Universality of kernel random matrices and kernel regression in the quadratic regime
Parthe Pandit, Zhichao Wang, Yizhe Zhu
Kernel Ridge Regression K$ Universality Random Matrix Kernel Regression Kernel Extension Positive Definite Kernel Parameterized Regime Linear Kernel

June 1, 2024

Conformal Transformation of Kernels: A Geometric Perspective on Text Classification
Ioana Rădulescu, Alexandra Băicoianu, Adela Mihai
Text Classification Small Kernel Gaussian Kernel Conformal Mapping Cosine Kernel Linear Kernel

January 17, 2024

Multiple Locally Linear Kernel Machines
David Picard
Linear Classifier Non Linear Kernel Learning Linear Kernel

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 Spectral Mixture Linear Kernel

March 28, 2023

LinK: Linear Kernel for LiDAR-based 3D Perception
Tao Lu, Xiang Ding, Haisong Liu, Gangshan Wu, Limin Wang
3D Object Detection 3D Perception 3D Semantic Segmentation 3D Object Detection Benchmark Tiltable Link LiDAR Based Perception Linear Kernel

February 7, 2023

Linear-scaling kernels for protein sequences and small molecules outperform deep learning while providing uncertainty quantitation and improved interpretability
Jonathan Parkinson, Wei Wang
Deep Learning Inherent Interpretability Gaussian Process Protein Sequence Small Molecule Linear Kernel Kernel Convolution

October 11, 2022

FusionDeepMF: A Dual Embedding based Deep Fusion Model for Recommendation
Supriyo Mandal, Abyayananda Maiti
Matrix Factorization Effective Recommendation User Item Traditional Recommender Dual System Linear Kernel

Linear Kernel

Papers

Universality of kernel random matrices and kernel regression in the quadratic regime

Conformal Transformation of Kernels: A Geometric Perspective on Text Classification

Multiple Locally Linear Kernel Machines

Sparsity-Aware Distributed Learning for Gaussian Processes with Linear Multiple Kernel

LinK: Linear Kernel for LiDAR-based 3D Perception

Linear-scaling kernels for protein sequences and small molecules outperform deep learning while providing uncertainty quantitation and improved interpretability

FusionDeepMF: A Dual Embedding based Deep Fusion Model for Recommendation