Edge-Based Hyperdimensional Learning System with Brain-Like Neural Adaptation

Authors: Zhuowen Zou (University of California, San Diego); Yeseong Kim (Daegu Institue of Science and Technology, South Korea); Farhad Imani (University of Connecticut); Haleh Alimohamadi (University of California, San Diego); Rosario Cammarota (Intel Labs); and Mohsen Imani (University of California, Irvine)

Abstract: Hyperdimensional Computing (HDC) is a brain-inspired learning approach for efficient and robust learning on today’s embedded devices. Encoding, or transforming the input data into high-dimensional representation, is the key first step of HDC before performing a learning task. In this paper, we have developed NeuralHD, a new HDC approach with a dynamic encoder for adaptive learning. Inspired by human neural regeneration study in neuroscience, NeuralHD identifies insignificant dimensions and regenerates those dimensions to enhance the learning capability and robustness. We also present a scalable learning framework to distribute NeuralHD computation over edge devices in IoT systems. Our solution enables edge devices capable of real-time learning from both labeled and unlabeled data. Our evaluation on a wide range of practical classification tasks shows that NeuralHD provides 5.7× and 6.1× (12.3× and 14.1×) faster and more energy-efficient training as compared to the HD-based algorithms (DNNs) running on the same platform.




Back to Technical Papers Archive Listing