Enhancing The Reliability of Out-of-distribution Image Detection in Neural Networks

TL;DR

This paper introduces ODIN, a simple method that improves out-of-distribution image detection in neural networks by using temperature scaling and input perturbations, achieving state-of-the-art results without modifying the original models.

Contribution

The paper presents ODIN, a novel technique that enhances OOD detection accuracy through a straightforward approach applicable to pre-trained networks.

Findings

ODIN significantly reduces false positive rates in OOD detection.

The method is compatible with various network architectures and datasets.

ODIN outperforms baseline methods by a large margin.

Abstract

We consider the problem of detecting out-of-distribution images in neural networks. We propose ODIN, a simple and effective method that does not require any change to a pre-trained neural network. Our method is based on the observation that using temperature scaling and adding small perturbations to the input can separate the softmax score distributions between in- and out-of-distribution images, allowing for more effective detection. We show in a series of experiments that ODIN is compatible with diverse network architectures and datasets. It consistently outperforms the baseline approach by a large margin, establishing a new state-of-the-art performance on this task. For example, ODIN reduces the false positive rate from the baseline 34.7% to 4.3% on the DenseNet (applied to CIFAR-10) when the true positive rate is 95%.