Uncertainty Quantification of Deep Learning for Spatiotemporal Data: Challenges and Opportunities
It tackles the problem of unreliable deep learning predictions in critical domains like disaster management and autonomous driving, but is incremental as it reviews existing methods and challenges.
This paper provides an overview of uncertainty quantification (UQ) in deep learning for spatiotemporal data, addressing the challenge of incorrect predictions with unwarranted confidence in high-stakes applications, and identifies future research directions.
With the advancement of GPS, remote sensing, and computational simulations, large amounts of geospatial and spatiotemporal data are being collected at an increasing speed. Such emerging spatiotemporal big data assets, together with the recent progress of deep learning technologies, provide unique opportunities to transform society. However, it is widely recognized that deep learning sometimes makes unexpected and incorrect predictions with unwarranted confidence, causing severe consequences in high-stake decision-making applications (e.g., disaster management, medical diagnosis, autonomous driving). Uncertainty quantification (UQ) aims to estimate a deep learning model's confidence. This paper provides a brief overview of UQ of deep learning for spatiotemporal data, including its unique challenges and existing methods. We particularly focus on the importance of uncertainty sources. We identify several future research directions for spatiotemporal data.