e-Distance Weighted Support Vector Regression

TL;DR

This paper introduces e-Distance Weighted Support Vector Regression (e-DWSVR), a novel method that improves robustness to noisy data and distributional differences in boundary data, scalable to large datasets.

Contribution

The paper proposes e-DWSVR, which optimizes both minimum margin and mean functional margin, and employs scalable optimization strategies for large-scale problems.

Findings

e-DWSVR outperforms existing methods on benchmark datasets.

The approach effectively handles noisy data and distributional shifts.

Scalable to large datasets using CD and ASGD strategies.

Abstract

We propose a novel support vector regression approach called e-Distance Weighted Support Vector Regression (e-DWSVR).e-DWSVR specifically addresses two challenging issues in support vector regression: first, the process of noisy data; second, how to deal with the situation when the distribution of boundary data is different from that of the overall data. The proposed e-DWSVR optimizes the minimum margin and the mean of functional margin simultaneously to tackle these two issues. In addition, we use both dual coordinate descent (CD) and averaged stochastic gradient descent (ASGD) strategies to make e-DWSVR scalable to large scale problems. We report promising results obtained by e-DWSVR in comparison with existing methods on several benchmark datasets.