Muffled Semi-Supervised Learning

TL;DR

This paper introduces a novel semi-supervised learning method where unlabeled data dampens the influence of labeled data, leading to improved AUC performance over traditional models.

Contribution

It proposes a new approach where unlabeled data muffle guidance, with variants demonstrating superior AUC compared to existing models.

Findings

Achieves higher AUC than boosted trees, random forests, and logistic regression.

Unlabeled data can be effectively used to improve model performance.

Experimental results validate the effectiveness of the muffling approach.

Abstract

We explore a novel approach to semi-supervised learning. This approach is contrary to the common approach in that the unlabeled examples serve to "muffle," rather than enhance, the guidance provided by the labeled examples. We provide several variants of the basic algorithm and show experimentally that they can achieve significantly higher AUC than boosted trees, random forests and logistic regression when unlabeled examples are available.