Enhancing The Reliability of Out-of-distribution Image Detection in Neural Networks
This addresses the reliability issue for neural network users in safety-critical applications like autonomous driving, though it is incremental as it builds on existing detection methods.
The paper tackles the problem of detecting out-of-distribution images in neural networks by proposing ODIN, a method that uses temperature scaling and input perturbations to improve detection, reducing the false positive rate from 34.7% to 4.3% on a DenseNet with CIFAR-10 at a 95% true positive rate.
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%.