Work

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

Year: 2021

Type: article

Abstract: , the claim that relevant out-distributions can lie in high likelihood regions of the data distribution, and that ood detection should be defined based on the data distribution's typical set. We highlight the consequences implied by assuming support overlap between in- and out-distributions, as well as the arbitrariness of the typical set for ood detection. Our results suggest that estimation error is a more plausible explanation than the misalignment between likelihood-based ood detection and out-distributions of interest, and we illustrate how even minimal estimation error can lead to ood detection failures, yielding implications for future work in deep generative modeling and ood detection.

Source: PubMed

DOI: 10.48550/arxiv.2107.06908

Authors: Zhang, Lily H., Goldstein, Mark, Ranganath, Rajesh

Institution: New York University

Language: English

Topic: Adversarial Robustness in Machine Learning

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funders: National Science Foundation, DeepMind, National Institutes of Health, National Heart, Lung, and Blood Institute

Awards: Deep probabilistic predictive models for stroke and coronary heart disease, G3329785188, TWC: Medium: Scaling proof-based verifiable computation, NRT-HDR: FUTURE Foundations, Translation, and Responsibility for Data Science Impact