HDnn-PIM: Efficient in Memory Design of Hyperdimensional Computing with Feature Extraction
HDnn-PIM: Efficient in Memory Design of Hyperdimensional Computing with Feature Extraction
复制标题
DOI：
10.1145/3526241.3530331
复制
发表时间：
2022-06
期刊：
影响因子：
--
通讯作者：
Arpan Dutta;Saransh Gupta;Behnam Khaleghi;Rishikanth Chandrasekaran;Weihong Xu;Tajana Simunic
中科院分区：
文献类型：
--
作者：
Arpan Dutta;Saransh Gupta;Behnam Khaleghi;Rishikanth Chandrasekaran;Weihong Xu;Tajana Simunic
脑启发的超维计算（HD）是一种新的机器学习方法，利用简单和高度并行化的操作。不幸的是，迄今为止，没有一个已发表的高清计算算法能够准确地分类更复杂的图像数据集，比如CIFAR100。在这项工作中，我们提出了hdn - pim，它通过使用内存处理来实现复杂图像的特征提取和基于高清的分类。我们比较了HDnn-PIM与HD-only和CNN在各种图像数据集上的实现。与纯高清计算相比，HDnn-PIM的准确率提高了52.4%。与最先进的cnn相比，它的准确率提高了1.2%，但内存占用减少了3.63倍，MAC操作减少了1.53倍。此外，HDnn-PIM比RTX 3090 GPU快3.6x-223倍，比最先进的FloatPIM节能3.7倍。
Brain-inspired Hyperdimensional (HD) computing is a new machine learning approach that leverages simple and highly parallelizable operations. Unfortunately, none of the published HD computing algorithms to date have been able to accurately classify more complex image datasets, such as CIFAR100. In this work, we propose HDnn-PIM, that implements both feature extraction and HD-based classification for complex images by using processing-in-memory. We compare HDnn-PIM with HD-only and CNN implementations for various image datasets. HDnn-PIM achieves 52.4% higher accuracy as compared to pure HD computing. It also gains 1.2% accuracy improvement over state-of-the-art CNNs, but with 3.63x smaller memory footprint and 1.53x less MAC operations. Furthermore, HDnn-PIM is 3.6x-223x faster than RTX 3090 GPU, and 3.7x more energy efficient than state-of-the-art FloatPIM.