Uncertainty Quantification for Deep Regression using Contextualised Normalizing Flows
This addresses uncertainty quantification for safe decision-making in high-risk domains, offering a practical post hoc solution without model modifications, though it is incremental as it builds on existing normalizing flow techniques.
The paper tackles the problem of quantifying uncertainty in deep regression models by introducing MCNF, a post hoc method that produces prediction intervals and full predictive distributions without retraining, and shows it is well-calibrated and competitive with state-of-the-art methods.
Quantifying uncertainty in deep regression models is important both for understanding the confidence of the model and for safe decision-making in high-risk domains. Existing approaches that yield prediction intervals overlook distributional information, neglecting the effect of multimodal or asymmetric distributions on decision-making. Similarly, full or approximated Bayesian methods, while yielding the predictive posterior density, demand major modifications to the model architecture and retraining. We introduce MCNF, a novel post hoc uncertainty quantification method that produces both prediction intervals and the full conditioned predictive distribution. MCNF operates on top of the underlying trained predictive model; thus, no predictive model retraining is needed. We provide experimental evidence that the MCNF-based uncertainty estimate is well calibrated, is competitive with state-of-the-art uncertainty quantification methods, and provides richer information for downstream decision-making tasks.