MIMHD: Accurate and Efficient Hyperdimensional Inference Using Multi-Bit In-Memory Computing
Authors:
- 26 July 2021
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Abstract
Hyperdimensional Computing (HDC) is an emerging computational framework that mimics important brain functions by operating over high-dimensional vectors, called hypervectors (HVs). In-memory computing implementations of HDC are desirable since they can significantly reduce data transfer overheads. All existing in-memory HDC platforms consider binary HVs where each dimension is represented with a single bit. However, utilizing multi-bit HVs allows HDC to achieve acceptable accuracies in lower dimensions which in turn leads to higher energy efficiencies. Thus, we propose a highly accurate and efficient multi-bit in-memory HDC inference platform called MIMHD. MIMHD supports multi-bit operations using ferroelectric field-effect transistor (FeFET) crossbar arrays for multiply-and-add and FeFET multi-bit content-addressable memories for associative search. We also introduce a novel hardware-aware retraining framework (HWART) that trains the HDC model to learn to work with MIMHD. For six popular datasets and 4000 dimension HVs, MIMHD using 3-bit (2-bit) precision HVs achieves (i) average accuracies of 92.6% (88.9%) which is 8.5% (4.8%) higher than binary implementations; (ii) 84.1× (78.6×) energy improvement over a GPU, and (iii) 38.4×(34.3×) speedup over a GPU, respectively. The 3-bit MIMHD is 4.3× and 13× faster and more energy-efficient than binary HDC accelerators while achieving similar accuracies.Keywords
All Related Versions
Funding Information
- Semiconductor Research Corporation
- Office of Naval Research
This publication has 19 references indexed in Scilit:
- Efficient Biosignal Processing Using Hyperdimensional Computing: Network Templates for Combined Learning and Classification of ExG SignalsProceedings of the IEEE, 2018
- A ferroelectric field effect transistor based synaptic weight cellJournal of Physics D: Applied Physics, 2018
- Exploring Hyperdimensional Associative Memory2017
- Nanoconnectomic upper bound on the variability of synaptic plasticityeLife, 2015
- Multi-column deep neural networks for image classification2012
- Introducing a New Benchmarked Dataset for Activity Monitoring2012
- Human Activity Recognition on Smartphones Using a Multiclass Hardware-Friendly Support Vector Machine2012
- Hyperdimensional Computing: An Introduction to Computing in Distributed Representation with High-Dimensional Random VectorsCognitive Computation, 2009
- Pruning Training Sets for Learning of Object Categories2005
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998