Out-of-distribution Detection and Generation using Soft Brownian Offset Sampling and Autoencoders

TL;DR

This paper introduces a novel method combining autoencoders and Soft Brownian Offset sampling to generate realistic out-of-distribution datasets, enhancing detection and validation in neural networks, especially for time series and trajectory data.

Contribution

The paper presents a new approach for generating out-of-distribution samples using autoencoders and a Soft Brownian Offset technique, improving OOD detection and model validation.

Findings

Enhanced OOD detection for MNIST dataset.

Effective generation of synthetic out-of-distribution time series.

Improved validation of trajectory prediction algorithms.

Abstract

Deep neural networks often suffer from overconfidence which can be partly remedied by improved out-of-distribution detection. For this purpose, we propose a novel approach that allows for the generation of out-of-distribution datasets based on a given in-distribution dataset. This new dataset can then be used to improve out-of-distribution detection for the given dataset and machine learning task at hand. The samples in this dataset are with respect to the feature space close to the in-distribution dataset and therefore realistic and plausible. Hence, this dataset can also be used to safeguard neural networks, i.e., to validate the generalization performance. Our approach first generates suitable representations of an in-distribution dataset using an autoencoder and then transforms them using our novel proposed Soft Brownian Offset method. After transformation, the decoder part of the autoencoder allows for the generation of these implicit out-of-distribution samples. This newly generated dataset then allows for mixing with other datasets and thus improved training of an out-of-distribution classifier, increasing its performance. Experimentally, we show that our approach is promising for time series using synthetic data. Using our new method, we also show in a quantitative case study that we can improve the out-of-distribution detection for the MNIST dataset. Finally, we provide another case study on the synthetic generation of out-of-distribution trajectories, which can be used to validate trajectory prediction algorithms for automated driving.