2021 IEEE International Conference on Acoustics, Speech and Signal Processing

6-11 June 2021 • Toronto, Ontario, Canada

Extracting Knowledge from Information

2021 IEEE International Conference on Acoustics, Speech and Signal Processing

6-11 June 2021 • Toronto, Ontario, Canada

Extracting Knowledge from Information

Technical Program

Paper Detail

Paper IDASPS-1.5
Paper Title Unsupervised Clustering of Time Series Signals using Neuromorphic Energy-Efficient Temporal Neural Networks
Authors Shreyas Chaudhari, Harideep Nair, Jose Moura, John Shen, Carnegie Mellon University, United States
SessionASPS-1: Architectures
LocationGather.Town
Session Time:Tuesday, 08 June, 16:30 - 17:15
Presentation Time:Tuesday, 08 June, 16:30 - 17:15
Presentation Poster
Topic Applied Signal Processing Systems: Signal Processing Hardware [DIS-PROG, DIS-MLTC, DIS-SOCP]
IEEE Xplore Open Preview  Click here to view in IEEE Xplore
Virtual Presentation  Click here to watch in the Virtual Conference
Abstract Unsupervised time series clustering is a challenging problem with diverse industrial applications such as anomaly detection, bio-wearables, etc. These applications typically involve small, low-power devices on the edge that collect and process real-time sensory signals. State-of-the-art time-series clustering methods perform some form of loss minimization that is extremely computationally intensive from the perspective of edge devices. In this work, we propose a neuromorphic approach to unsupervised time series clustering based on Temporal Neural Networks that is capable of ultra lowpower, continuous online learning. We demonstrate its clustering performance on a subset of UCR Time Series Archive datasets. Our results show that the proposed approach either outperforms or performs similarly to most of the existing algorithms while being far more amenable for efficient hardware implementation. Our hardware assessment analysis shows that in 7 nm CMOS the proposed architecture, on average, consumes only about 0.005 mm squared die area and 22 micro watts power and can process each signal with about 5 ns latency.