Skip to main content

Competitive Normalised Least Squares Regression.

Waqas, J. and Bouchachia, A., 2021. Competitive Normalised Least Squares Regression. IEEE Transactions on Neural Networks and Learning Systems, 32 (7), 3262-3267.

Full text available as:

Competitive_Normalized_Least-Squares_Regression.pdf - Published Version
Available under License Creative Commons Attribution.

[img] PDF
ieee_onlsr.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Attribution Non-commercial.


DOI: 10.1109/TNNLS.2020.3009777


Online learning has witnessed an increasing interest over the recent past due to its low computational requirements and its relevance to a broad range of streaming applications. In this paper, we focus on online regularised regression. We propose a novel efficient online regression algorithm, called Online Normalised Least Squares (ONLS). We perform theoretical analysis, by comparing the total loss of ONLS against the Normalised Gradient Descent algorithm (NGD) and the best offline LS predictor. We show in particular that ONLS allows for a better bias-variance tradeoff than those state-of-the-art gradient descent-based LS algorithms as well as a better control on the level of shrinkage of the features towards null. Finally, we conduct an empirical study to illustrate the great performance of ONLS against some state-of-the-art algorithms using real-world data.

Item Type:Article
Additional Information:Authors are supported by the European Commission under the Horizon 2020 Grant 687691 for the project PROTEUS
Uncontrolled Keywords:Competitive analysis; least-squares; prediction
Group:Faculty of Science & Technology
ID Code:34331
Deposited By: Symplectic RT2
Deposited On:27 Jul 2020 09:54
Last Modified:14 Mar 2022 14:23


Downloads per month over past year

More statistics for this item...
Repository Staff Only -