Feature Encoding with AutoEncoders for Weakly-supervised Anomaly Detection
This work addresses the problem of limited labeled anomaly data for researchers and practitioners in anomaly detection, representing an incremental improvement with a novel encoding approach.
The paper tackles weakly-supervised anomaly detection by proposing a novel feature encoding strategy using autoencoders, which leverages hidden representation, reconstruction residual vector, and reconstruction error to improve detection performance, achieving superior results over competitive methods.
Weakly-supervised anomaly detection aims at learning an anomaly detector from a limited amount of labeled data and abundant unlabeled data. Recent works build deep neural networks for anomaly detection by discriminatively mapping the normal samples and abnormal samples to different regions in the feature space or fitting different distributions. However, due to the limited number of annotated anomaly samples, directly training networks with the discriminative loss may not be sufficient. To overcome this issue, this paper proposes a novel strategy to transform the input data into a more meaningful representation that could be used for anomaly detection. Specifically, we leverage an autoencoder to encode the input data and utilize three factors, hidden representation, reconstruction residual vector, and reconstruction error, as the new representation for the input data. This representation amounts to encode a test sample with its projection on the training data manifold, its direction to its projection and its distance to its projection. In addition to this encoding, we also propose a novel network architecture to seamlessly incorporate those three factors. From our extensive experiments, the benefits of the proposed strategy are clearly demonstrated by its superior performance over the competitive methods.