Multitask Learning of Vegetation Biochemistry from Hyperspectral Data

TL;DR

This paper introduces a multitask Gaussian process approach to improve vegetation biochemical prediction from hyperspectral data, especially when ground truth data is limited, by leveraging related biochemical information.

Contribution

It presents a novel Gaussian process multitask learning method that models inter-biochemical relationships with multiple covariance functions, enhancing prediction accuracy with scarce data.

Findings

Outperformed existing methods on two real-world datasets.

Effective in scenarios with limited ground truth data.

Utilizes inter-relationship modeling between biochemicals.

Abstract

Statistical models have been successful in accurately estimating the biochemical contents of vegetation from the reflectance spectra. However, their performance deteriorates when there is a scarcity of sizable amount of ground truth data for modeling the complex non-linear relationship occurring between the spectrum and the biochemical quantity. We propose a novel Gaussian process based multitask learning method for improving the prediction of a biochemical through the transfer of knowledge from the learned models for predicting related biochemicals. This method is most advantageous when there are few ground truth data for the biochemical of interest, but plenty of ground truth data for related biochemicals. The proposed multitask Gaussian process hypothesizes that the inter-relationship between the biochemical quantities is better modeled by using a combination of two or more covariance functions and inter-task correlation matrices. In the experiments, our method outperformed the current methods on two real-world datasets.