Likelihood-based Out-of-Distribution Detection with Denoising Diffusion Probabilistic Models
Joseph S Goodier (University of Bath),* Neill Campbell (University of Bath)
The 34th British Machine Vision Conference
Abstract
Out-of-Distribution detection between dataset pairs has been extensively explored with generative models. We show that likelihood-based Out-of-Distribution detection can be extended to diffusion models by leveraging the fact that they, like other likelihood-based generative models, are dramatically affected by the input sample complexity. Currently, all Out-of-Distribution detection methods with Diffusion Models are reconstruction-based. We propose a new likelihood ratio for Out-of-Distribution detection with Deep Denoising Diffusion Models, which we call the Complexity Corrected Likelihood Ratio. Our likelihood ratio is constructed using Evidence Lower-Bound evaluations from an individual model at various noising levels. We present results that are comparable to state-of-the-art Out-of-Distribution detection methods with generative models.Citation
@inproceedings{Goodier_2023_BMVC,
author = {Joseph S Goodier and Neill Campbell},
title = {Likelihood-based Out-of-Distribution Detection with Denoising Diffusion Probabilistic Models},
booktitle = {34th British Machine Vision Conference 2023, {BMVC} 2023, Aberdeen, UK, November 20-24, 2023},
publisher = {BMVA},
year = {2023},
url = {https://bmvc2022.mpi-inf.mpg.de/BMVC2023/0094.pdf}
}
Copyright © 2023
The British Machine Vision Conference is organised by