Hyperdimensional Feature Fusion for Out-Of-Distribution Detection

TL;DR

This paper proposes a novel hyperdimensional feature fusion method for out-of-distribution detection that leverages similarity-preserving projections and class-specific descriptors across multiple neural network layers, outperforming existing methods.

Contribution

It introduces a hyperdimensional computing approach to fuse multi-layer features for improved OOD detection, a novel application in this domain.

Findings

Outperforms state-of-the-art OOD detection methods.

Hyperdimensional fusion of multiple layers enhances detection accuracy.

Simple cosine similarity effectively identifies OOD samples.

Abstract

We introduce powerful ideas from Hyperdimensional Computing into the challenging field of Out-of-Distribution (OOD) detection. In contrast to most existing work that performs OOD detection based on only a single layer of a neural network, we use similarity-preserving semi-orthogonal projection matrices to project the feature maps from multiple layers into a common vector space. By repeatedly applying the bundling operation $\oplus$, we create expressive class-specific descriptor vectors for all in-distribution classes. At test time, a simple and efficient cosine similarity calculation between descriptor vectors consistently identifies OOD samples with better performance than the current state-of-the-art. We show that the hyperdimensional fusion of multiple network layers is critical to achieve best general performance.