HybridFlow: Quantification of Aleatoric and Epistemic Uncertainty with a Single Hybrid Model
This addresses a key problem in Bayesian deep learning for high-stakes applications by providing a robust framework, though it appears incremental as it builds on existing uncertainty quantification methods.
The paper tackles the challenge of quantifying both aleatoric and epistemic uncertainty in machine learning by introducing HybridFlow, a modular hybrid architecture that unifies these uncertainties in a single model, showing improved calibration and alignment with model error across regression tasks like depth estimation and ice sheet emulation.
Uncertainty quantification is critical for ensuring robustness in high-stakes machine learning applications. We introduce HybridFlow, a modular hybrid architecture that unifies the modeling of aleatoric and epistemic uncertainty by combining a Conditional Masked Autoregressive normalizing flow for estimating aleatoric uncertainty with a flexible probabilistic predictor for epistemic uncertainty. The framework supports integration with any probabilistic model class, allowing users to easily adapt HybridFlow to existing architectures without sacrificing predictive performance. HybridFlow improves upon previous uncertainty quantification frameworks across a range of regression tasks, such as depth estimation, a collection of regression benchmarks, and a scientific case study of ice sheet emulation. We also provide empirical results of the quantified uncertainty, showing that the uncertainty quantified by HybridFlow is calibrated and better aligns with model error than existing methods for quantifying aleatoric and epistemic uncertainty. HybridFlow addresses a key challenge in Bayesian deep learning, unifying aleatoric and epistemic uncertainty modeling in a single robust framework.