Out-of-distribution Detection via Frequency-regularized Generative Models

TL;DR

This paper introduces a frequency-regularized training framework for deep generative models to improve out-of-distribution detection, addressing their tendency to rely on background information and achieve state-of-the-art results.

Contribution

The authors propose a novel frequency-regularized learning (FRL) method that incorporates high-frequency information into training to enhance OOD detection across various generative architectures.

Findings

FRL improves OOD detection performance significantly.

FRL achieves 147× faster inference speed.

FRL maintains high image generation quality.

Abstract

Modern deep generative models can assign high likelihood to inputs drawn from outside the training distribution, posing threats to models in open-world deployments. While much research attention has been placed on defining new test-time measures of OOD uncertainty, these methods do not fundamentally change how deep generative models are regularized and optimized in training. In particular, generative models are shown to overly rely on the background information to estimate the likelihood. To address the issue, we propose a novel frequency-regularized learning FRL framework for OOD detection, which incorporates high-frequency information into training and guides the model to focus on semantically relevant features. FRL effectively improves performance on a wide range of generative architectures, including variational auto-encoder, GLOW, and PixelCNN++. On a new large-scale evaluation task, FRL achieves the state-of-the-art performance, outperforming a strong baseline Likelihood Regret by 10.7% (AUROC) while achieving 147$\times$ faster inference speed. Extensive ablations show that FRL improves the OOD detection performance while preserving the image generation quality. Code is available at https://github.com/mu-cai/FRL.