Projects per year
Abstract
Kernel recursive least squares (KRLS) is a widely used online machine learning algorithm for time series predictions. In this article, we present the mixed-precision KRLS, producing equivalent prediction accuracy to double-precision KRLS with a higher training throughput and a lower memory footprint. The mixed-precision KRLS applies single-precision arithmetic to the computation components being not only numerically resilient but also computationally intensive. Our mixed-precision KRLS demonstrates the 1.32, 1.15, 1.29, 1.09, and 1.08x training throughput improvements using 24.95%, 24.74%, 24.89%, 24.48%, and 24.20% less memory footprint without losing any prediction accuracy compared to double-precision KRLS for a 3-D nonlinear regression, a Lorenz chaotic time series, a Mackey-Glass chaotic time series, a sunspot number time series, and a sea surface temperature time series, respectively.
Original language | English |
---|---|
Journal | IEEE Transactions on Neural Networks and Learning Systems |
Early online date | 16 Dec 2020 |
DOIs | |
Publication status | Early online date - 16 Dec 2020 |
Fingerprint
Dive into the research topics of 'Mixed-Precision Kernel Recursive Least Squares'. Together they form a unique fingerprint.-
R6584CSC: Energy Efficient Trransprecision Techniques for Linear system Solvers
Vandierendonck, H. (PI)
09/04/2018 → …
Project: Research
-
R6551CSC: Open TransPREcision COMPuting
Woods, R. (PI), Karakonstantis, G. (CoI) & Vandierendonck, H. (CoI)
03/11/2016 → 31/12/2020
Project: Research