DisCoPatch: Taming Adversarially-driven Batch Statistics for Improved Out-of-Distribution Detection

TL;DR

DisCoPatch introduces an unsupervised adversarial VAE framework that leverages batch statistics to improve out-of-distribution detection, especially under covariate shifts, achieving state-of-the-art results efficiently.

Contribution

The paper proposes DisCoPatch, a novel method that exploits batch statistics in adversarial discriminators with BN for enhanced covariate shift OOD detection.

Findings

Achieves 95.5% AUROC on ImageNet-1K(-C)

Outperforms prior methods on Near-OOD benchmarks

Operates with a compact 25MB model at low latency

Abstract

Out-of-distribution (OOD) detection holds significant importance across many applications. While semantic and domain-shift OOD problems are well-studied, this work focuses on covariate shifts - subtle variations in the data distribution that can degrade machine learning performance. We hypothesize that detecting these subtle shifts can improve our understanding of in-distribution boundaries, ultimately improving OOD detection. In adversarial discriminators trained with Batch Normalization (BN), real and adversarial samples form distinct domains with unique batch statistics - a property we exploit for OOD detection. We introduce DisCoPatch, an unsupervised Adversarial Variational Autoencoder (VAE) framework that harnesses this mechanism. During inference, batches consist of patches from the same image, ensuring a consistent data distribution that allows the model to rely on batch statistics. DisCoPatch uses the VAE's suboptimal outputs (generated and reconstructed) as negative samples to train the discriminator, thereby improving its ability to delineate the boundary between in-distribution samples and covariate shifts. By tightening this boundary, DisCoPatch achieves state-of-the-art results in public OOD detection benchmarks. The proposed model not only excels in detecting covariate shifts, achieving 95.5% AUROC on ImageNet-1K(-C) but also outperforms all prior methods on public Near-OOD (95.0%) benchmarks. With a compact model size of 25MB, it achieves high OOD detection performance at notably lower latency than existing methods, making it an efficient and practical solution for real-world OOD detection applications. The code is publicly available.