Mixed-Precision Kernel Recursive Least Squares

JunKyu Lee,Dimitrios S Nikolopoulos,Hans Vandierendonck

IEEE Transactions on Neural Networks and Learning Systems（2022）

引用 1|浏览44

暂无评分

摘要

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 n...

查看译文

关键词

Kernel,Training,Prediction algorithms,Throughput,Support vector machines,Time series analysis,Memory management

AI 理解论文

溯源树

样例

生成溯源树，研究论文发展脉络

Chat Paper

正在生成论文摘要