Robust Task Learning Based on Nonlinear Regression With Mixtures of Student-<italic>t</italic> Distributions

Robust Task Learning Based on Nonlinear Regression With Mixtures of Student-<italic>t</italic> Distributions

We propose a robust task learning method based on nonlinear regression model with mixtures of t-distributions. The model can adaptively reduce the effects of complex noises and accurately learn the nonlinear structure of targets. By introducing latent variables, the model is expressed into a hierarc...

Full description

Bibliographic Details
Main Authors: Chunzheng Cao, Ziyue Wang, Jian Qing Shi, Yunjie Chen
Format: Article
Language:English
Published: IEEE 2020-01-01
Series:IEEE Access
Subjects:
Task learning
robust nonlinear regression
outlier
EM algorithm
kernel method
Online Access:https://ieeexplore.ieee.org/document/9113244/

Internet

https://ieeexplore.ieee.org/document/9113244/

Similar Items