Shaken, and Stirred: Long-Range Dependencies Enable Robust Outlier Detection with PixelCNN++

TL;DR

This paper introduces simple, computationally inexpensive transformations to improve outlier detection in deep generative models by emphasizing long-range dependencies, achieving state-of-the-art results on natural images.

Contribution

It proposes two families of bijective transformations, 'stirring' and 'shaking', that mitigate local bias effects in PixelCNN++ likelihoods, enhancing outlier detection performance.

Findings

Transformations improve outlier detection accuracy.

Methods generalize to other generative models.

Approaches outperform existing techniques on complex datasets.

Abstract

Reliable outlier detection is critical for real-world deployment of deep learning models. Although extensively studied, likelihoods produced by deep generative models have been largely dismissed as being impractical for outlier detection. First, deep generative model likelihoods are readily biased by low-level input statistics. Second, many recent solutions for correcting these biases are computationally expensive, or do not generalize well to complex, natural datasets. Here, we explore outlier detection with a state-of-the-art deep autoregressive model: PixelCNN++. We show that biases in PixelCNN++ likelihoods arise primarily from predictions based on local dependencies. We propose two families of bijective transformations -- ``stirring'' and ``shaking'' -- which ameliorate low-level biases and isolate the contribution of long-range dependencies to PixelCNN++ likelihoods. These transformations are inexpensive and readily computed at evaluation time. We test our approaches extensively with five grayscale and six natural image datasets and show that they achieve or exceed state-of-the-art outlier detection, particularly on datasets with complex, natural images. We also show that our solutions work well with other types of generative models (generative flows and variational autoencoders) and that their efficacy is governed by each model's reliance on local dependencies. In sum, lightweight remedies suffice to achieve robust outlier detection on image data with deep generative models.