Further Analysis of Outlier Detection with Deep Generative Models

10/25/2020
by   Ziyu Wang, et al.
6

The recent, counter-intuitive discovery that deep generative models (DGMs) can frequently assign a higher likelihood to outliers has implications for both outlier detection applications as well as our overall understanding of generative modeling. In this work, we present a possible explanation for this phenomenon, starting from the observation that a model's typical set and high-density region may not conincide. From this vantage point we propose a novel outlier test, the empirical success of which suggests that the failure of existing likelihood-based outlier tests does not necessarily imply that the corresponding generative model is uncalibrated. We also conduct additional experiments to help disentangle the impact of low-level texture versus high-level semantics in differentiating outliers. In aggregate, these results suggest that modifications to the standard evaluation practices and benchmarks commonly applied in the literature are needed.

READ FULL TEXT

page 6

page 12

page 17

08/29/2022

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

Reliable outlier detection is critical for real-world applications of de...
06/15/2021

Robust Out-of-Distribution Detection on Deep Probabilistic Generative Models

Out-of-distribution (OOD) detection is an important task in machine lear...
07/14/2021

Understanding Failures in Out-of-Distribution Detection with Deep Generative Models

Deep generative models (DGMs) seem a natural fit for detecting out-of-di...
09/22/2021

Entropic Issues in Likelihood-Based OOD Detection

Deep generative models trained by maximum likelihood remain very popular...
10/26/2018

Outlier Detection using Generative Models with Theoretical Performance Guarantees

This paper considers the problem of recovering signals from compressed m...
08/19/2021

Efficient remedies for outlier detection with variational autoencoders

Deep networks often make confident, yet incorrect, predictions when test...
07/29/2020

Evaluation of Sampling Methods for Scatterplots

Given a scatterplot with tens of thousands of points or even more, a nat...