Log In Sign Up

A statistical theory of out-of-distribution detection

by   Xi Wang, et al.

We introduce a principled approach to detecting out-of-distribution (OOD) data by exploiting a connection to data curation. In data curation, we exclude ambiguous or difficult-to-classify input points from the dataset, and these excluded points are by definition OOD. We can therefore obtain the likelihood for OOD points by using a principled generative model of data-curation initially developed to explain the cold-posterior effect in Bayesian neural networks (Aitchison 2020). This model gives higher OOD probabilities when predictive uncertainty is higher and can be trained using maximum-likelihood jointly over the in-distribution and OOD points. This approach gives superior performance to past methods that did not provide a probability for OOD points, and therefore could not be trained using maximum-likelihood.


page 2

page 5


A statistical theory of cold posteriors in deep neural networks

To get Bayesian neural networks to perform comparably to standard neural...

Investigating maximum likelihood based training of infinite mixtures for uncertainty quantification

Uncertainty quantification in neural networks gained a lot of attention ...

Stochastic Maximum Likelihood Optimization via Hypernetworks

This work explores maximum likelihood optimization of neural networks th...

Amortized Conditional Normalized Maximum Likelihood

While deep neural networks provide good performance for a range of chall...

Entropic Issues in Likelihood-Based OOD Detection

Deep generative models trained by maximum likelihood remain very popular...

Towards Unifying Logical Entailment and Statistical Estimation

This paper gives a generative model of the interpretation of formal logi...

Data augmentation in Bayesian neural networks and the cold posterior effect

Data augmentation is a highly effective approach for improving performan...