Likelihood Regret: An Out-of-Distribution Detection Score For Variational Auto-encoder

TL;DR

This paper introduces Likelihood Regret, a new out-of-distribution detection score specifically designed for Variational Auto-encoders, addressing limitations of existing likelihood-based methods.

Contribution

The paper proposes Likelihood Regret as an effective OOD detection score for VAEs, outperforming existing approaches in empirical benchmarks.

Findings

Likelihood Regret achieves superior OOD detection performance on VAEs.

Existing likelihood-based methods often fail for VAEs in OOD detection.

Empirical results demonstrate the effectiveness of the proposed method.

Abstract

Deep probabilistic generative models enable modeling the likelihoods of very high dimensional data. An important application of generative modeling should be the ability to detect out-of-distribution (OOD) samples by setting a threshold on the likelihood. However, some recent studies show that probabilistic generative models can, in some cases, assign higher likelihoods on certain types of OOD samples, making the OOD detection rules based on likelihood threshold problematic. To address this issue, several OOD detection methods have been proposed for deep generative models. In this paper, we make the observation that many of these methods fail when applied to generative models based on Variational Auto-encoders (VAE). As an alternative, we propose Likelihood Regret, an efficient OOD score for VAEs. We benchmark our proposed method over existing approaches, and empirical results suggest that our method obtains the best overall OOD detection performances when applied to VAEs.