<ul><li>Continual learning (CL) focuses on learning a sequence of tasks incrementally.</li><li>This paper addresses the challenges of class-incremental learning (CIL), including catastrophic forgetting and inter-task class separation.</li><li>The proposed method, Kernel Linear Discriminant Analysis (KLDA), utilizes features learned in a foundation model (FM) to overcome these challenges.</li><li>KLDA incorporates the Radial Basis Function (RBF) kernel and its Random Fourier Features (RFF) to improve feature representations and achieves comparable performance to joint training of all classes.</li></ul>

Continual Learning Using a Kernel-Based Method Over Foundation Models

Discover more