Subset Scanning Over Neural Network Activations
This work addresses the critical issue of anomaly detection for neural networks, particularly for adversarial noise, but it is incremental as it adapts existing methods to a new domain.
The paper tackles the problem of detecting anomalous inputs in neural networks by applying subset scanning methods to node activations, achieving detection of adversarial noise on CIFAR-10 with a log-linear time search guarantee.
This work views neural networks as data generating systems and applies anomalous pattern detection techniques on that data in order to detect when a network is processing an anomalous input. Detecting anomalies is a critical component for multiple machine learning problems including detecting adversarial noise. More broadly, this work is a step towards giving neural networks the ability to recognize an out-of-distribution sample. This is the first work to introduce "Subset Scanning" methods from the anomalous pattern detection domain to the task of detecting anomalous input of neural networks. Subset scanning treats the detection problem as a search for the most anomalous subset of node activations (i.e., highest scoring subset according to non-parametric scan statistics). Mathematical properties of these scoring functions allow the search to be completed in log-linear rather than exponential time while still guaranteeing the most anomalous subset of nodes in the network is identified for a given input. Quantitative results for detecting and characterizing adversarial noise are provided for CIFAR-10 images on a simple convolutional neural network. We observe an "interference" pattern where anomalous activations in shallow layers suppress the activation structure of the original image in deeper layers.