SHEAR er
Authors:
Open Access
- 10 August 2020
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 241-246
Abstract
Hyperdimensional computing (HD) is an emerging paradigm for machine learning based on the evidence that the brain computes on high-dimensional, distributed, representations of data. The main operation of HD is encoding, which transfers the input data to hyperspace by mapping each input feature to a hypervector, followed by a bundling procedure that adds up the hypervectors to realize the encoding hypervector. The operations of HD are simple and highly parallelizable, but the large number of operations hampers the efficiency of HD in embedded domain. In this paper, we propose SHEARer, an algorithmhardware co-optimization to improve the performance and energy consumption of HD computing. We gain insight from a prudent scheme of approximating the hypervectors that, thanks to error resiliency of HD, has minimal impact on accuracy while provides high prospect for hardware optimization. Unlike previous works that generate the encoding hypervectors in full precision and then and then perform ex-post quantization, we compute the encoding hypervectors in an approximate manner that saves resources yet affords high accuracy. We also propose a novel FPGA architecture that achieves striking performance through massive parallelism with low power consumption. Moreover, we develop a software framework that enables training HD models by emulating the proposed approximate encodings. The FPGA implementation of SHEARer achieves an average throughput boost of 104,904× (15.7×) and energy savings of up to 56,044× (301×) compared to state-of-the-art encoding methods implemented on Raspberry Pi 3 (GeForce GTX 1080 Ti) using practical machine learning datasets.Keywords
Funding Information
- SRC Global Research Collaboration (GRC)
- NSF (1911095, 2003279)
- DARPA
This publication has 8 references indexed in Scilit:
- Hardware Optimizations of Dense Binary Hyperdimensional Computing: Rematerialization of Hypervectors, Binarized Bundling, and Combinational Associative MemoryACM Journal on Emerging Technologies in Computing Systems, 2019
- An Introduction to Hyperdimensional Computing for RoboticsKI - Künstliche Intelligenz, 2019
- A Programmable Hyper-Dimensional Processor Architecture for Human-Centric IoTIEEE Journal on Emerging and Selected Topics in Circuits and Systems, 2019
- Learning sensorimotor control with neuromorphic sensors: Toward hyperdimensional active perceptionScience Robotics, 2019
- Early Olfactory Processing in Drosophila: Mechanisms and PrinciplesAnnual Review of Neuroscience, 2013
- Hyperdimensional Computing: An Introduction to Computing in Distributed Representation with High-Dimensional Random VectorsCognitive Computation, 2009
- Olfactory Representations byDrosophilaMushroom Body NeuronsJournal of Neurophysiology, 2008
- Holographic reduced representationsIEEE Transactions on Neural Networks, 1995