InQMAD: Incremental Quantum Measurement Anomaly Detection

TL;DR

InQMAD introduces an incremental quantum-inspired anomaly detection method capable of continuous density estimation in streaming data, effectively handling concept drift and outliers with constant update complexity.

Contribution

The paper proposes a novel quantum measurement-based incremental anomaly detection approach that outperforms existing methods in streaming data scenarios.

Findings

Outperforms 12 state-of-the-art algorithms on multiple datasets

Handles concept drift and outliers effectively

Maintains constant update complexity O(1)

Abstract

Streaming anomaly detection refers to the problem of detecting anomalous data samples in streams of data. This problem poses challenges that classical and deep anomaly detection methods are not designed to cope with, such as conceptual drift and continuous learning. State-of-the-art flow anomaly detection methods rely on fixed memory using hash functions or nearest neighbors that may not be able to constrain high-frequency values as in a moving average or remove seamless outliers and cannot be trained in an end-to-end deep learning architecture. We present a new incremental anomaly detection method that performs continuous density estimation based on random Fourier features and the mechanism of quantum measurements and density matrices that can be viewed as an exponential moving average density. It can process potentially endless data and its update complexity is constant $O(1)$. A systematic evaluation against 12 state-of-the-art streaming anomaly detection algorithms using 12 streaming datasets is presented.