VARADE: a Variational-based AutoRegressive model for Anomaly Detection on the Edge

TL;DR

VARADE is a lightweight variational autoregressive model designed for real-time anomaly detection on edge devices, balancing accuracy, power consumption, and inference speed in industrial settings.

Contribution

This paper introduces VARADE, a novel variational-based autoregressive model optimized for edge computing, addressing the computational limitations of existing deep learning solutions.

Findings

Outperforms state-of-the-art algorithms in anomaly detection accuracy

Achieves lower power consumption on edge platforms

Maintains high inference frequency suitable for real-time applications

Abstract

Detecting complex anomalies on massive amounts of data is a crucial task in Industry 4.0, best addressed by deep learning. However, available solutions are computationally demanding, requiring cloud architectures prone to latency and bandwidth issues. This work presents VARADE, a novel solution implementing a light autoregressive framework based on variational inference, which is best suited for real-time execution on the edge. The proposed approach was validated on a robotic arm, part of a pilot production line, and compared with several state-of-the-art algorithms, obtaining the best trade-off between anomaly detection accuracy, power consumption and inference frequency on two different edge platforms.