PQ-HDC: Projection-Based Quantization Scheme for Flexible and Efficient Hyperdimensional Computing
Journal
IFIP Advances in Information and Communication Technology
Journal Volume
627
Pages
425-435
ISBN
9.78303E+12
Date Issued
2021
Author(s)
Abstract
Brain-inspired Hyperdimensional (HD) computing is an emerging technique for low-power/energy designs in many machine learning tasks. Recent works further exploit the low-cost quantized (bipolarized or ternarized) HD model and report dramatic improvements in energy efficiency. However, the quantization loss of HD models leads to a severe drop in classification accuracy. This paper proposes a projection-based quantization framework for HD computing (PQ-HDC) to achieve a flexible and efficient trade-off between accuracy and efficiency. While previous works exploit thresholding-quantization schemes, the proposed PQ-HDC progressively reduces quantization loss using a linear combination of bipolarized HD models. Furthermore, PQ-HDC allows quantization with flexible bit-width while preserving the computational efficiency of the Hamming distance computation. Experimental results on the benchmark dataset demonstrate that PQ-HDC achieves a 2.82% improvement in accuracy over the state-of-the-art method. © 2021, IFIP International Federation for Information Processing.
Subjects
Brain-inspired computing; Dynamic model; Energy efficiency; Hyperdimensional Computing
SDGs
Other Subjects
Computational efficiency; Economic and social effects; Energy efficiency; Hamming distance; Benchmark datasets; Classification accuracy; Hamming distance computation; Linear combinations; Low power/energy; Quantization loss; Quantization schemes; State-of-the-art methods; Artificial intelligence
Type
conference paper