Online nonlinear estimation via iterative L2-space projections: Reproducing Kernel of subspace

Motoya Ohnishi, Masahiro Yukawa

Research output: Contribution to journalArticlepeer-review


We propose a novel online learning paradigm for nonlinear-function estimation tasks based on the iterative projections in the L2space with probability measure reflecting the stochastic property of input signals. The proposed learning algorithm exploits the reproducing kernel of the so-called dictionary subspace, based on the fact that any finite-dimensional space of functions has a reproducing kernel characterized by the Gram matrix. The L2-space geometry provides the best decorrelation property in principle. The proposed learning paradigm is significantly different from the conventional kernelbased learning paradigm in two senses: (i) the whole space is not a reproducing kernel Hilbert space and (ii) the minimum mean squared error estimator gives the best approximation of the desired nonlinear function in the dictionary subspace. It preserves efficiency in computing the inner product as well as in updating the Gram matrix when the dictionary grows. Monotone approximation, asymptotic optimality, and convergence of the proposed algorithm are analyzed based on the variable-metric version of adaptive projected subgradient method. Numerical examples show the efficacy of the proposed algorithm for real data over a variety of methods including the extended Kalman filter and many batch machine-learning methods such as the multilayer perceptron.

Original languageEnglish
JournalUnknown Journal
Publication statusPublished - 2017 Dec 12


  • Kernel adaptive filter
  • Lspace
  • Metric projection
  • Online learning
  • Recursive least squares

ASJC Scopus subject areas

  • General

Fingerprint Dive into the research topics of 'Online nonlinear estimation via iterative L<sup>2</sup>-space projections: Reproducing Kernel of subspace'. Together they form a unique fingerprint.

Cite this