The Effect of Optimization Methods on the Robustness of Out-of-Distribution Detection Approaches

TL;DR

This paper investigates how different optimization methods affect the robustness of out-of-distribution detection approaches in deep neural networks, revealing that optimization choices significantly influence detection performance.

Contribution

It introduces a robustness score to evaluate the impact of optimization methods on OOD detection and demonstrates that certain optimization techniques enhance robustness.

Findings

Optimization methods significantly influence OOD detection robustness.

Some optimization techniques yield better solutions for OOD detection.

The proposed robustness score enables fair comparison of OOD approaches.

Abstract

Deep neural networks (DNNs) have become the de facto learning mechanism in different domains. Their tendency to perform unreliably on out-of-distribution (OOD) inputs hinders their adoption in critical domains. Several approaches have been proposed for detecting OOD inputs. However, existing approaches still lack robustness. In this paper, we shed light on the robustness of OOD detection (OODD) approaches by revealing the important role of optimization methods. We show that OODD approaches are sensitive to the type of optimization method used during training deep models. Optimization methods can provide different solutions to a non-convex problem and so these solutions may or may not satisfy the assumptions (e.g., distributions of deep features) made by OODD approaches. Furthermore, we propose a robustness score that takes into account the role of optimization methods. This provides a sound way to compare OODD approaches. In addition to comparing several OODD approaches using our proposed robustness score, we demonstrate that some optimization methods provide better solutions for OODD approaches.