Online efficient learning with quantized KLMS and L1 regularization

@article{Chen2012OnlineEL,
  title={Online efficient learning with quantized KLMS and L1 regularization},
  author={Badong Chen and Songlin Zhao and Sohan Seth and Jos{\'e} Carlos Pr{\'i}ncipe},
  journal={The 2012 International Joint Conference on Neural Networks (IJCNN)},
  year={2012},
  pages={1-6}
}
In a recent work, we have proposed the quantized kernel least mean square (QKLMS) algorithm, which is quite effective in online learning sequentially a nonlinear mapping with a slowly growing radial basis function (RBF) structure. In this paper, in order to further reduce the network size, we propose a sparse QKLMS algorithm, which is derived by adding a sparsity inducing l1 norm penalty of the coefficients to the squared error cost. Simulation examples show that the new algorithm works… CONTINUE READING
Highly Cited
This paper has 29 citations. REVIEW CITATIONS

Citations

Publications citing this paper.
Showing 1-10 of 15 extracted citations

Efficient Dictionary-Refining Kernel Adaptive Filter With Fundamental Insights

IEEE Transactions on Signal Processing • 2016
View 14 Excerpts
Highly Influenced

An efficient sparse kernel adaptive filtering algorithm based on isomorphism between functional subspace and Euclidean space

2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) • 2014
View 12 Excerpts
Highly Influenced

Online Dictionary Learning for Kernel LMS

IEEE Transactions on Signal Processing • 2014
View 4 Excerpts
Highly Influenced

Budget Online Learning Algorithm for Least Squares SVM

IEEE Transactions on Neural Networks and Learning Systems • 2017
View 1 Excerpt

On initial convergence behavior of the kernel least mean square algorithm

2015 International Joint Conference on Neural Networks (IJCNN) • 2015
View 1 Excerpt

References

Publications referenced by this paper.
Showing 1-8 of 8 references

Sparsity regularized recursive least squares adaptive filtering,

E. M. Eksioglu
IET Signal Process., • 2011
View 5 Excerpts
Highly Influenced

$l_{0}$ Norm Constraint LMS Algorithm for Sparse System Identification

IEEE Signal Processing Letters • 2009
View 5 Excerpts
Highly Influenced

The kernel recursive least-squares algorithm

IEEE Transactions on Signal Processing • 2004
View 6 Excerpts
Highly Influenced

Quantized Kernel Least Mean Square Algorithm

IEEE Transactions on Neural Networks and Learning Systems • 2012
View 4 Excerpts

Sparse LMS for system identification

2009 IEEE International Conference on Acoustics, Speech and Signal Processing • 2009

The Kernel Least-Mean-Square Algorithm

IEEE Transactions on Signal Processing • 2008
View 2 Excerpts

Similar Papers

Loading similar papers…