Improving Interpretability of Scores in Anomaly Detection Based on Gaussian-Bernoulli Restricted Boltzmann Machine

arXiv:2403.12672v1 Announce Type: new
Abstract: Gaussian-Bernoulli restricted Boltzmann machines (GBRBMs) are often used for semi-supervised anomaly detection, where they are trained using only normal data points. In GBRBM-based anomaly detection, normal and anomalous data are classified based on a score that is identical to an energy function of the marginal GBRBM. However, the classification threshold is difficult to set to an appropriate value, as this score cannot be interpreted. In this study, we propose a measure that improves score's interpretability …

abstract anomaly anomaly detection arxiv boltzmann boltzmann machines cs.ai cs.lg data detection energy function interpretability machine machines normal semi-supervised stat.ml type