Target Detection on Hyperspectral Images Using MCMC and VI Trained Bayesian Neural Networks

TL;DR

This paper compares MCMC and variational inference trained Bayesian neural networks for target detection in hyperspectral images, highlighting their performance and the importance of uncertainty quantification in high-stakes applications.

Contribution

It demonstrates the application and comparison of MCMC and VI trained BNNs for hyperspectral target detection, emphasizing the impact of training methods on results.

Findings

Both MCMC and VI BNNs perform well in target detection.

Different training methods yield different results for the same model.

Sufficient computational resources should guide the choice of training method.

Abstract

Neural networks (NN) have become almost ubiquitous with image classification, but in their standard form produce point estimates, with no measure of confidence. Bayesian neural networks (BNN) provide uncertainty quantification (UQ) for NN predictions and estimates through the posterior distribution. As NN are applied in more high-consequence applications, UQ is becoming a requirement. BNN provide a solution to this problem by not only giving accurate predictions and estimates, but also an interval that includes reasonable values within a desired probability. Despite their positive attributes, BNN are notoriously difficult and time consuming to train. Traditional Bayesian methods use Markov Chain Monte Carlo (MCMC), but this is often brushed aside as being too slow. The most common method is variational inference (VI) due to its fast computation, but there are multiple concerns with its efficacy. We apply and compare MCMC- and VI-trained BNN in the context of target detection in hyperspectral imagery (HSI), where materials of interest can be identified by their unique spectral signature. This is a challenging field, due to the numerous permuting effects practical collection of HSI has on measured spectra. Both models are trained using out-of-the-box tools on a high fidelity HSI target detection scene. Both MCMC- and VI-trained BNN perform well overall at target detection on a simulated HSI scene. This paper provides an example of how to utilize the benefits of UQ, but also to increase awareness that different training methods can give different results for the same model. If sufficient computational resources are available, the best approach rather than the fastest or most efficient should be used, especially for high consequence problems.