Multivariate Deep Evidential Regression

Abstract

There is significant need for principled uncertainty reasoning in machine learning systems as they are increasingly deployed in safety-critical domains. A new approach with uncertainty-aware neural networks (NNs), based on learning evidential distributions for aleatoric and epistemic uncertainties, shows promise over traditional deterministic methods and typical Bayesian NNs, yet several important gaps in the theory and implementation of these networks remain. We discuss three issues with a proposed solution to extract aleatoric and epistemic uncertainties from regression-based neural networks. The approach derives a technique by placing evidential priors over the original Gaussian likelihood function and training the NN to infer the hyperparameters of the evidential distribution. Doing so allows for the simultaneous extraction of both uncertainties without sampling or utilization of out-of-distribution data for univariate regression tasks. We describe the outstanding issues in detail, provide a possible solution, and generalize the deep evidential regression technique for multivariate cases.

20 pages, 13 figures

References in corpus (10)

• What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?
• On Calibration of Modern Neural Networks
• Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shift
• Evidential Deep Learning to Quantify Classification Uncertainty
• Loss Surfaces, Mode Connectivity, and Fast Ensembling of DNNs
• Noise Contrastive Priors for Functional Uncertainty
• Classifier-agnostic saliency map extraction
• Towards calibrated and scalable uncertainty representations for neural networks
• Sampling-based Bayesian Inference with gradient uncertainty
• A Parsimonious Tour of Bayesian Model Uncertainty