Kernel Recursive Least Squares Dictionary Learning Algorithm
By: Ghasem Alipoor, Karl Skretting
Potential Business Impact:
Teaches computers to learn patterns faster.
We propose an efficient online dictionary learning algorithm for kernel-based sparse representations. In this framework, input signals are nonlinearly mapped to a high-dimensional feature space and represented sparsely using a virtual dictionary. At each step, the dictionary is updated recursively using a novel algorithm based on the recursive least squares (RLS) method. This update mechanism works with single samples or mini-batches and maintains low computational complexity. Experiments on four datasets across different domains show that our method not only outperforms existing online kernel dictionary learning approaches but also achieves classification accuracy close to that of batch-trained models, while remaining significantly more efficient.
Similar Papers
Online multidimensional dictionary learning
Numerical Analysis
Improves how computers understand complex data.
Topological Dictionary Learning
Signal Processing
Finds hidden patterns in connected data.
Models Parametric Analysis via Adaptive Kernel Learning
Optimization and Control
Finds best settings for computer models.