Mind the Nuisance: Gaussian Process Classification using Privileged Noise

TL;DR

This paper introduces GPC+ which incorporates privileged information as noise in Gaussian Process classifiers, improving classification accuracy by leveraging additional knowledge during training.

Contribution

The novel approach models privileged information as noise in GPC, enhancing confidence estimation and outperforming existing methods like SVM+ and standard GPC.

Findings

GPC+ outperforms standard GPC and SVM+ on public datasets.

Privileged information can be effectively modeled as noise in Gaussian Processes.

Deep learning models can be compressed as privileged information.

Abstract

The learning with privileged information setting has recently attracted a lot of attention within the machine learning community, as it allows the integration of additional knowledge into the training process of a classifier, even when this comes in the form of a data modality that is not available at test time. Here, we show that privileged information can naturally be treated as noise in the latent function of a Gaussian Process classifier (GPC). That is, in contrast to the standard GPC setting, the latent function is not just a nuisance but a feature: it becomes a natural measure of confidence about the training data by modulating the slope of the GPC sigmoid likelihood function. Extensive experiments on public datasets show that the proposed GPC method using privileged noise, called GPC+, improves over a standard GPC without privileged knowledge, and also over the current state-of-the-art SVM-based method, SVM+. Moreover, we show that advanced neural networks and deep learning methods can be compressed as privileged information.