SearcHD: A Memory-Centric Hyperdimensional Computing With Stochastic Training
SearcHD: A Memory-Centric Hyperdimensional Computing With Stochastic Training
复制标题
DOI：
10.1109/tcad.2019.2952544
复制
发表时间：
2020-10-01
影响因子：
2.9
通讯作者：
Rosing, Tajana
中科院分区：
文献类型：
--
作者：
Imani, Mohsen;Yin, Xunzhao;Rosing, Tajana
大脑启发的超维度（HD）计算通过计算长二进制向量（也称为超向量）来模拟认知任务，而不是用数字计算。然而，我们观察到，为了在实际应用中提供可接受的分类精度，HD算法需要在非二元超向量上进行训练和测试。在本文中，我们提出了一个完全二值化的HD计算算法SearcHD。SearcHD将每个数据点映射到具有二进制元素的高维空间。SearcHD实现了一种全二进制训练方法，该方法为每个类生成多个二进制超向量，而不是用非二进制元素训练HD模型。我们还利用非易失性存储器（nvm）的模拟特性在存储器中执行所有编码、训练和推理计算。我们在广泛的分类应用中评估了SearcHD的效率和准确性。我们的评估表明，与最先进的高清计算算法相比，SearcHD可以提供平均高31.1倍的能源效率和12.8倍的训练速度。
Brain-inspired hyperdimensional (HD) computing emulates cognitive tasks by computing with long binary vectors-also know as hypervectors-as opposed to computing with numbers. However, we observed that in order to provide acceptable classification accuracy on practical applications, HD algorithms need to be trained and tested on nonbinary hypervectors. In this article, we propose SearcHD, a fully binarized HD computing algorithm with a fully binary training. SearcHD maps every data points to a high-dimensional space with binary elements. Instead of training an HD model with nonbinary elements, SearcHD implements a full binary training method which generates multiple binary hypervectors for each class. We also use the analog characteristic of nonvolatile memories (NVMs) to perform all encoding, training, and inference computations in memory. We evaluate the efficiency and accuracy of SearcHD on a wide range of classification applications. Our evaluation shows that SearcHD can provide on average 31.1\x higher energy efficiency and 12.8x faster training as compared to the state-of-the-art HD computing algorithms.