Distributional Gaussian Process Layers for Outlier Detection in Image Segmentation

Abstract

We propose a parameter efficient Bayesian layer for hierarchical convolutional Gaussian Processes that incorporates Gaussian Processes operating in Wasserstein-2 space to reliably propagate uncertainty. This directly replaces convolving Gaussian Processes with a distance-preserving affine operator on distributions. Our experiments on brain tissue-segmentation show that the resulting architecture approaches the performance of well-established deterministic segmentation algorithms (U-Net), which has never been achieved with previous hierarchical Gaussian Processes. Moreover, by applying the same segmentation model to out-of-distribution data (i.e., images with pathology such as brain tumors), we show that our uncertainty estimates result in out-of-distribution detection that outperforms the capabilities of previous Bayesian networks and reconstruction-based approaches that learn normative distributions.

References in corpus (7)

• On Calibration of Modern Neural Networks
• Uncertainty Estimation Using a Single Deep Deterministic Neural Network
• Bayesian Deep Learning and a Probabilistic Perspective of Generalization
• Stochastic Segmentation Networks: Modelling Spatially Correlated Aleatoric Uncertainty
• One Versus all for deep Neural Network Incertitude (OVNNI) quantification
• Revisiting One-vs-All Classifiers for Predictive Uncertainty and Out-of-Distribution Detection in Neural Networks
• Hierarchical Gaussian Processes with Wasserstein-2 Kernels