Anomaly Detection via Gumbel Noise Score Matching
This work addresses anomaly detection for categorical and image data, presenting a novel method that is incremental in its adaptation of score matching techniques.
The paper tackles anomaly detection in categorical and image data by proposing Gumbel Noise Score Matching (GNSM), which estimates gradients of log likelihoods for continuously relaxed categorical distributions, achieving consistently high performance on tabular datasets and effectively identifying poor segmentation predictions in images with strong correlation to ground-truth metrics.
We propose Gumbel Noise Score Matching (GNSM), a novel unsupervised method to detect anomalies in categorical data. GNSM accomplishes this by estimating the scores, i.e. the gradients of log likelihoods w.r.t.~inputs, of continuously relaxed categorical distributions. We test our method on a suite of anomaly detection tabular datasets. GNSM achieves a consistently high performance across all experiments. We further demonstrate the flexibility of GNSM by applying it to image data where the model is tasked to detect poor segmentation predictions. Images ranked anomalous by GNSM show clear segmentation failures, with the outputs of GNSM strongly correlating with segmentation metrics computed on ground-truth. We outline the score matching training objective utilized by GNSM and provide an open-source implementation of our work.