Multiscale Score Matching for Out-of-Distribution Detection
This addresses the challenge of out-of-distribution detection in computer vision, which is crucial for reliable AI systems, and it is incremental as it builds on existing score-based methods.
The paper tackles the problem of detecting out-of-distribution images by using norms of score estimates at multiple noise scales, and it significantly outperforms state-of-the-art methods, such as effectively separating CIFAR-10 and SVHN images.
We present a new methodology for detecting out-of-distribution (OOD) images by utilizing norms of the score estimates at multiple noise scales. A score is defined to be the gradient of the log density with respect to the input data. Our methodology is completely unsupervised and follows a straight forward training scheme. First, we train a deep network to estimate scores for levels of noise. Once trained, we calculate the noisy score estimates for N in-distribution samples and take the L2-norms across the input dimensions (resulting in an NxL matrix). Then we train an auxiliary model (such as a Gaussian Mixture Model) to learn the in-distribution spatial regions in this L-dimensional space. This auxiliary model can now be used to identify points that reside outside the learned space. Despite its simplicity, our experiments show that this methodology significantly outperforms the state-of-the-art in detecting out-of-distribution images. For example, our method can effectively separate CIFAR-10 (inlier) and SVHN (OOD) images, a setting which has been previously shown to be difficult for deep likelihood models.