Feature space approximation for kernel-based supervised learning
This method offers a way to make kernel-based supervised learning more efficient and generalizable for practitioners dealing with large datasets and high-dimensional features.
This paper introduces a method to approximate high-dimensional feature vectors in supervised learning, aiming to reduce training data size, storage, and computational complexity. The method also acts as a regularization technique, demonstrating significant improvements over using the full training dataset in classification and regression tasks.
We propose a method for the approximation of high- or even infinite-dimensional feature vectors, which play an important role in supervised learning. The goal is to reduce the size of the training data, resulting in lower storage consumption and computational complexity. Furthermore, the method can be regarded as a regularization technique, which improves the generalizability of learned target functions. We demonstrate significant improvements in comparison to the computation of data-driven predictions involving the full training data set. The method is applied to classification and regression problems from different application areas such as image recognition, system identification, and oceanographic time series analysis.