Efficient KLMS and KRLS Algorithms: A Random Fourier Feature Perspective

Autor:	Bouboulis, Pantelis, Pougkakiotis, Spyridon, Theodoridis, Sergios
Rok vydání:	2016
Předmět:	Computer Science - Learning Statistics - Machine Learning K.3.2 I.5.4
Druh dokumentu:	Working Paper
Popis:	We present a new framework for online Least Squares algorithms for nonlinear modeling in RKH spaces (RKHS). Instead of implicitly mapping the data to a RKHS (e.g., kernel trick), we map the data to a finite dimensional Euclidean space, using random features of the kernel's Fourier transform. The advantage is that, the inner product of the mapped data approximates the kernel function. The resulting "linear" algorithm does not require any form of sparsification, since, in contrast to all existing algorithms, the solution's size remains fixed and does not increase with the iteration steps. As a result, the obtained algorithms are computationally significantly more efficient compared to previously derived variants, while, at the same time, they converge at similar speeds and to similar error floors. Comment: presented in the 2016 IEEE Workshop on Statistical Signal Processing (SSP 16)
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/1606.03685 Zobrazit plný text záznamu View this record from Arxiv