How to get the most out of Twinned Regression Methods

TL;DR

This paper investigates twinned regression methods that predict differences between targets, exploring their algorithmic components, ensemble quality, hybrid models with k-NN, and a semi-supervised approach to improve regression accuracy.

Contribution

It introduces a detailed analysis of twinned regression algorithms, combines neural networks with k-NN, and proposes a semi-supervised scheme for enhanced regression performance.

Findings

Decomposition of algorithm steps reveals their impact on accuracy.

Hybrid neural network and k-NN method improves efficiency.

Semi-supervised approach enhances regression performance.

Abstract

Twinned regression methods are designed to solve the dual problem to the original regression problem, predicting differences between regression targets rather then the targets themselves. A solution to the original regression problem can be obtained by ensembling predicted differences between the targets of an unknown data point and multiple known anchor data points. We explore different aspects of twinned regression methods: (1) We decompose different steps in twinned regression algorithms and examine their contributions to the final performance, (2) We examine the intrinsic ensemble quality, (3) We combine twin neural network regression with k-nearest neighbor regression to design a more accurate and efficient regression method, and (4) we develop a simplified semi-supervised regression scheme.