Twin neural network regression is a semi-supervised regression algorithm. Issue 4 (1st December 2022)

• Record Type: Journal Article
• Authors: Wetzel, Sebastian J; Melko, Roger G; Tamblyn, Isaac
• Is Part Of: Machine learning: science and technology. Volume 3:Issue 4(2022)
• Publication Date: 2022-12-01
• Subjects: artificial neural networks -- regression -- semi-supervised learning; 006.31
• DOI: 10.1088/2632-2153/ac9885
• Languages: English
• ISSNs: 2632-2153

Abstract

Abstract: Twin neural network regression (TNNR) is trained to predict differences between the target values of two different data points rather than the targets themselves. By ensembling predicted differences between the targets of an unseen data point and all training data points, it is possible to obtain a very accurate prediction for the original regression problem. Since any loop of predicted differences should sum to zero, loops can be supplied to the training data, even if the data points themselves within loops are unlabelled. Semi-supervised training improves TNNR performance, which is already state of the art, significantly.