Reliable Neural Networks for Regression Uncertainty Estimation

Abstract

While deep neural networks are highly performant and successful in a wide range of real-world problems, estimating their predictive uncertainty remains a challenging task. To address this challenge, we propose and implement a loss function for regression uncertainty estimation based on the Bayesian Validation Metric (BVM) framework while using ensemble learning. The proposed loss reproduces maximum likelihood estimation in the limiting case. A series of experiments on in-distribution data show that the proposed method is competitive with existing state-of-the-art methods. Experiments on out-of-distribution data show that the proposed method is robust to statistical change and exhibits superior predictive capability.

References in corpus (2)

• Probabilistic Backpropagation for Scalable Learning of Bayesian Neural Networks
• Maximum Likelihood Uncertainty Estimation: Robustness to Outliers

Cited by in corpus (2)

• Calibration in Machine Learning Uncertainty Quantification: beyond consistency to target adaptivity
• Insurance pricing with hierarchically structured data: An illustration with a workers' compensation insurance portfolio