A new algorithm for support vector regression with automatic selection of hyperparameters

作者：

Highlights：

• Establishing an extended primal objective function based on probability regularization leading to a data dependent proportion parameter ν and ϵ.

• This regularization establishes the equivalence of the ϵ-SVR and ν-SVR in which ν is specified as a proportion of ϵ−1log(1+ϵ).

• It proposes values that are equivalent to the maximum likelihood estimates under the distributional assumptions by the loss function; ν is an explicit function of ϵ.

摘要

•Establishing an extended primal objective function based on probability regularization leading to a data dependent proportion parameter ν and ϵ.•This regularization establishes the equivalence of the ϵ-SVR and ν-SVR in which ν is specified as a proportion of ϵ−1log(1+ϵ).•It proposes values that are equivalent to the maximum likelihood estimates under the distributional assumptions by the loss function; ν is an explicit function of ϵ.

论文关键词：Automatic selection,Loss functions,Noise models,Parameter estimation,Probability regularization

论文评审过程：Received 20 December 2021, Revised 19 July 2022, Accepted 16 August 2022, Available online 18 August 2022, Version of Record 26 August 2022.

论文官网地址：https://doi.org/10.1016/j.patcog.2022.108989