Contrastive Learning for Regression on Hyperspectral Data

TL;DR

This paper proposes a contrastive learning framework for the regression tasks for hyperspectral data, provides a collection of transformations relevant for augmenting hyperspectral data, and investigates contrastive learning for regression.

Abstract

Contrastive learning has demonstrated great effectiveness in representation learning especially for image classification tasks. However, there is still a shortage in the studies targeting regression tasks, and more specifically applications on hyperspectral data. In this paper, we propose a contrastive learning framework for the regression tasks for hyperspectral data. To this end, we provide a collection of transformations relevant for augmenting hyperspectral data, and investigate contrastive learning for regression. Experiments on synthetic and real hyperspectral datasets show that the proposed framework and transformations significantly improve the performance of regression models, achieving better scores than other state-of-the-art transformations.

Figures (3)

• Figure 1: Architecture of the proposed method.
• Figure 2: Original and Transformed Examples from Real Data.
• Figure 3: Prediction Error Distribution.