Uncertainty Tube Visualization of Particle Trajectories

TL;DR

This paper presents the uncertainty tube, a new visualization method that effectively and efficiently depicts uncertainty in neural network-predicted particle trajectories, enhancing interpretability in scientific applications.

Contribution

The paper introduces the uncertainty tube, a novel visualization technique that captures nonsymmetric uncertainty in particle trajectories predicted by neural networks.

Findings

The uncertainty tube accurately visualizes uncertainty in synthetic datasets.

It effectively conveys uncertainty in simulation data.

The method integrates multiple uncertainty quantification techniques.

Abstract

Predicting particle trajectories with neural networks (NNs) has substantially enhanced many scientific and engineering domains. However, effectively quantifying and visualizing the inherent uncertainty in predictions remains challenging. Without an understanding of the uncertainty, the reliability of NN models in applications where trustworthiness is paramount is significantly compromised. This paper introduces the uncertainty tube, a novel, computationally efficient visualization method designed to represent this uncertainty in NN-derived particle paths. Our key innovation is the design and implementation of a superelliptical tube that accurately captures and intuitively conveys nonsymmetric uncertainty. By integrating well-established uncertainty quantification techniques, such as Deep Ensembles, Monte Carlo Dropout (MC Dropout), and Stochastic Weight Averaging-Gaussian (SWAG), we demonstrate the practical utility of the uncertainty tube, showcasing its application on both synthetic and simulation datasets.