Dec 6, 2018 · We propose incorporating the Mahalanobis distance in latent space to better capture these out-of-distribution samples and our results show that this method ...

Dec 6, 2018 · We propose incorporating the Mahalanobis distance in latent space to better capture these out-of-distribution samples and our results show that ...

This paper proposes an efficient data augmentation network to detect out-of-distribution image data by introducing a set of common geometric operations into ...

Dec 30, 2022 · Bibliographic details on Improving Reconstruction Autoencoder Out-of-distribution Detection with Mahalanobis Distance.

Improving reconstruction autoencoder out-of-distribution detection with mahalanobis distance. T Denouden, R Salay, K Czarnecki, V Abdelzad, B Phan, S Vernekar.

In [7] the OoD detection meth- ods of Mahalanobis distance and autoencoder were merged into a unified framework, supposing that the latter could be thus ...

For OOD detection, we use two metrics un- der both scenarios to compute Scorecla and Scorerec in latent space, i.e., Mahalanobis distance and Euclidean dis-.

Deep learning for anomaly detection: A survey. 2019. 1. Improving reconstruction autoencoder out-of-distribution detection with mahalanobis distance. Jan 2018.

A PyTorch implementation of an autoencoder trained to minimize the Mahalanobis distance between input and reconstruction.

Dec 6, 2018 · Improving Reconstruction Autoencoder Out-of-distribution Detection with Mahalanobis Distance · Bayesian Variational Autoencoders for Unsupervised ...