ADIC: Anomaly Detection Integrated Circuit in 65nm CMOS utilizing Approximate Computing

TL;DR

This paper introduces a low-power anomaly detection integrated circuit (ADIC) in 65nm CMOS that uses approximate computing and ensemble neural networks to achieve significant energy savings while maintaining detection accuracy.

Contribution

It presents a novel ADIC design employing ensemble neural networks and approximate computing techniques for energy-efficient anomaly detection in hardware.

Findings

Achieves 18.5x energy reduction compared to full-precision computing.

Enables dynamic neural network sizing for energy-efficient operation.

Maintains high detection accuracy with reduced energy consumption.

Abstract

In this paper, we present a low-power anomaly detection integrated circuit (ADIC) based on a one-class classifier (OCC) neural network. The ADIC achieves low-power operation through a combination of (a) careful choice of algorithm for online learning and (b) approximate computing techniques to lower average energy. In particular, online pseudoinverse update method (OPIUM) is used to train a randomized neural network for quick and resource efficient learning. An additional 42% energy saving can be achieved when a lighter version of OPIUM method is used for training with the same number of data samples lead to no significant compromise on the quality of inference. Instead of a single classifier with large number of neurons, an ensemble of K base learner approach is chosen to reduce learning memory by a factor of K. This also enables approximate computing by dynamically varying the neural network size based on anomaly detection. Fabricated in 65nm CMOS, the ADIC has K = 7 Base Learners (BL) with 32 neurons in each BL and dissipates 11.87pJ/OP and 3.35pJ/OP during learning and inference respectively at Vdd = 0.75V when all 7 BLs are enabled. Further, evaluated on the NASA bearing dataset, approximately 80% of the chip can be shut down for 99% of the lifetime leading to an energy efficiency of 0.48pJ/OP, an 18.5 times reduction over full-precision computing running at Vdd = 1.2V throughout the lifetime.