Kernel Dictionary Learning

Kernel Dictionary Learning (KDL) is a machine learning technique that extends traditional dictionary learning by using kernel methods to learn nonlinear representations of data. Current research focuses on improving KDL's efficiency for large datasets, often through reduced-kernel approximations and optimized algorithms like variations of AK-SVD, aiming to reduce computational cost while maintaining accuracy. These advancements are impacting various applications, including anomaly detection and image segmentation, by enabling the effective analysis of complex, high-dimensional data.

Papers

October 17, 2023

MRI brain tumor segmentation using informative feature vectors and kernel dictionary learning
Seyedeh Mahya Mousavi, Mohammad Mostafavi
Brain Tumor Segmentation Magnetic Resonance Image Feature Vector Informative Latent Representation Kernel Dictionary Learning

July 17, 2023

Anomaly Detection with Selective Dictionary Learning
Denis C. Ilie-Ablachim, Bogdan Dumitrescu
Anomaly Detection Outlier Detection Dictionary Learning Kernel Dictionary Learning
Reduced Kernel Dictionary Learning
Denis C. Ilie-Ablachim, Bogdan Dumitrescu
Sparse Representation Dictionary Learning Nonlinear Representation Kernel Based Representation Kernel Dictionary Learning
Classification with Incoherent Kernel Dictionary Learning
Denis C. Ilie-Ablachim, Bogdan Dumitrescu
Classification Code Individual Representation Singular Value Decomposition Dictionary Learning Classification Problem Kernel Dictionary Learning

Kernel Dictionary Learning

Papers

MRI brain tumor segmentation using informative feature vectors and kernel dictionary learning

Anomaly Detection with Selective Dictionary Learning

Reduced Kernel Dictionary Learning

Classification with Incoherent Kernel Dictionary Learning