StocHD: Stochastic Hyperdimensional System for Efficient and Robust Learning from Raw Data
StocHD: Stochastic Hyperdimensional System for Efficient and Robust Learning from Raw Data
复制标题
DOI：
10.1109/dac18074.2021.9586166
复制
发表时间：
2021-12
期刊：
影响因子：
--
通讯作者：
Prathyush Poduval;Zhuowen Zou;M. Najafi;H. Homayoun;M. Imani
中科院分区：
文献类型：
--
作者：
Prathyush Poduval;Zhuowen Zou;M. Najafi;H. Homayoun;M. Imani
超维计算（HDC）是一种神经启发的计算模型，它基于对人类大脑对高维数据表示（称为超向量）的操作的观察。尽管HDC在抽象信息的推理和关联方面非常强大，但在图像/视频等复杂数据的特征提取方面却很弱。因此，大多数现有的HDC解决方案依赖于昂贵的预处理算法进行特征提取。在本文中，我们提出了一种新的端到端超维系统，支持对原始数据的准确、高效和鲁棒学习。与之前使用HDC进行学习任务的工作不同，StocHD通过在HDC超向量上定义随机算法，将HDC功能扩展到计算领域。StocHD使整个学习应用程序（包括特征提取器）使用HDC数据表示进行处理，实现统一、高效、鲁棒和高度并行的计算。我们还提出了一种新颖的全数字化和可扩展的内存处理（PIM）架构，该架构利用HDC以内存为中心的特性来支持广泛的并行计算。我们对大范围分类任务的评估表明，平均而言，随机分散度提供了3.3倍和6.4倍（52.3倍和143倍）。Sx)与运行在PIM （NVIDIA GPU）上的最先进的HDC算法相比，速度更快，能效更高，同时提供16倍的计算鲁棒性。
Hyperdimensional Computing (HDC) is a neurally-inspired computation model working based on the observation that the human brain operates on high-dimensional representations of data, called hypervector. Although HDC is significantly powerful in reasoning and association of the abstract information, it is weak on features extraction from complex data such as image/video. As a result, most existing HDC solutions rely on expensive pre-processing algorithms for feature extraction. In this paper, we propose StocHD, a novel end-to-end hyperdimensional system that supports accurate, efficient, and robust learning over raw data. Unlike prior work that used HDC for learning tasks, StocHD expands HDC functionality to the computing area by mathematically defining stochastic arithmetic over HDC hypervectors. StocHD enables an entire learning application (including feature extractor) to process using HDC data representation, enabling uniform, efficient, robust, and highly parallel computation. We also propose a novel fully digital and scalable Processing In-Memory (PIM) architecture that exploits the HDC memory-centric nature to support extensively parallel computation. Our evaluation over a wide range of classification tasks shows that StocHD provides, on average, 3.3x and 6.4x (52.3x and 143.Sx) faster and higher energy efficiency as compared to state-of-the-art HDC algorithm running on PIM (NVIDIA GPU), while providing 16x higher computational robustness.