R2-AD2: Detecting Anomalies by Analysing the Raw Gradient

TL;DR

R2-AD2 is a semi-supervised anomaly detection method that analyzes the raw gradient distribution over training steps using an RNN to identify point anomalies across various domains.

Contribution

It introduces a novel approach that leverages raw gradient analysis with an RNN, enabling effective anomaly detection without domain-specific features.

Findings

Reliable detection of point anomalies in semi-supervised settings

Applicable across diverse domains without domain-dependent features

Data-driven approach using raw gradients and RNN architecture

Abstract

Neural networks follow a gradient-based learning scheme, adapting their mapping parameters by back-propagating the output loss. Samples unlike the ones seen during training cause a different gradient distribution. Based on this intuition, we design a novel semi-supervised anomaly detection method called R2-AD2. By analysing the temporal distribution of the gradient over multiple training steps, we reliably detect point anomalies in strict semi-supervised settings. Instead of domain dependent features, we input the raw gradient caused by the sample under test to an end-to-end recurrent neural network architecture. R2-AD2 works in a purely data-driven way, thus is readily applicable in a variety of important use cases of anomaly detection.