Enhanced ${q}$-Least Mean Square
Optimization and Control
2018-01-03 v1 Systems and Control
Other Statistics
Abstract
In this work, a new class of stochastic gradient algorithm is developed based on -calculus. Unlike the existing -LMS algorithm, the proposed approach fully utilizes the concept of -calculus by incorporating time-varying parameter. The proposed enhanced -LMS (-LMS) algorithm utilizes a novel, parameterless concept of error-correlation energy and normalization of signal to ensure high convergence, stability and low steady-state error. The proposed algorithm automatically adapts the learning rate with respect to the error. For the evaluation purpose the system identification problem is considered. Extensive experiments show better performance of the proposed -LMS algorithm compared to the standard -LMS approach.
Cite
@article{arxiv.1801.00410,
title = {Enhanced ${q}$-Least Mean Square},
author = {Shujaat Khan and Alishba Sadiq and Imran Naseem and Roberto Togneri and Mohammed Bennamoun},
journal= {arXiv preprint arXiv:1801.00410},
year = {2018}
}