Unsupervised Energy-based Out-of-distribution Detection using Stiefel-Restricted Kernel Machine

TL;DR

This paper introduces an unsupervised energy-based out-of-distribution detection method using Stiefel-Restricted Kernel Machines, which outperforms existing energy-based and generative models on standard datasets.

Contribution

It proposes a novel unsupervised OOD detection approach leveraging St-RKM with multiple energy functions, improving over prior energy-based detectors.

Findings

Outperforms existing energy-based OOD detectors

Demonstrates effectiveness on standard datasets

Ablation studies validate each energy function's utility

Abstract

Detecting out-of-distribution (OOD) samples is an essential requirement for the deployment of machine learning systems in the real world. Until now, research on energy-based OOD detectors has focused on the softmax confidence score from a pre-trained neural network classifier with access to class labels. In contrast, we propose an unsupervised energy-based OOD detector leveraging the Stiefel-Restricted Kernel Machine (St-RKM). Training requires minimizing an objective function with an autoencoder loss term and the RKM energy where the interconnection matrix lies on the Stiefel manifold. Further, we outline multiple energy function definitions based on the RKM framework and discuss their utility. In the experiments on standard datasets, the proposed method improves over the existing energy-based OOD detectors and deep generative models. Through several ablation studies, we further illustrate the merit of each proposed energy function on the OOD detection performance.