Distributional Uncertainty for Out-of-Distribution Detection

TL;DR

This paper introduces the Free-Energy Posterior Network, a novel approach that models distributional uncertainty for improved out-of-distribution detection and segmentation in deep neural networks, using a Beta distribution-based density estimator.

Contribution

It proposes a new framework that jointly models distributional uncertainty and OoD detection using free energy, with a Beta distribution density estimator and a loss for direct uncertainty estimation.

Findings

Effective OoD detection on Fishyscapes, RoadAnomaly, and Segment-Me-If-You-Can benchmarks.

Outperforms traditional energy-based methods in uncertainty estimation.

Enables learning OoD regions without post-hoc thresholding.

Abstract

Estimating uncertainty from deep neural networks is a widely used approach for detecting out-of-distribution (OoD) samples, which typically exhibit high predictive uncertainty. However, conventional methods such as Monte Carlo (MC) Dropout often focus solely on either model or data uncertainty, failing to align with the semantic objective of OoD detection. To address this, we propose the Free-Energy Posterior Network, a novel framework that jointly models distributional uncertainty and identifying OoD and misclassified regions using free energy. Our method introduces two key contributions: (1) a free-energy-based density estimator parameterized by a Beta distribution, which enables fine-grained uncertainty estimation near ambiguous or unseen regions; and (2) a loss integrated within a posterior network, allowing direct uncertainty estimation from learned parameters without requiring stochastic sampling. By integrating our approach with the residual prediction branch (RPL) framework, the proposed method goes beyond post-hoc energy thresholding and enables the network to learn OoD regions by leveraging the variance of the Beta distribution, resulting in a semantically meaningful and computationally efficient solution for uncertainty-aware segmentation. We validate the effectiveness of our method on challenging real-world benchmarks, including Fishyscapes, RoadAnomaly, and Segment-Me-If-You-Can.