Robust Deep Reservoir Computing Through Reliable Memristor With Improved Heat Dissipation Capability
Robust Deep Reservoir Computing Through Reliable Memristor With Improved Heat Dissipation Capability
复制标题
DOI：
10.1109/tcad.2020.3002539
复制
发表时间：
2021-03-01
影响因子：
2.9
通讯作者：
Yi, Yang
中科院分区：
文献类型：
--
作者：
An, Hongyu;Al-Mamun, Mohammad Shah;Yi, Yang
深度神经网络 (DNN) 是一种受大脑启发的学习方法，在执行推理任务之前需要大量数据进行训练。最近的研究表明，推理精度与 DNN 和数据集的大小之间存在很强的正相关性，这导致了对大型 DNN 的不可避免的需求。然而，传统的存储技术不足以应对数据集和神经网络大小的急剧增长。最近，电阻忆阻器由于其高密度和低功耗而被广泛认为是下一代存储器件。然而，其高开关电阻变化（周期间）限制了其在深度学习中的可行性。在这项工作中，制造并评估了一种具有增强散热功能的新型忆阻器配置，以应对这一挑战。我们的实验结果表明，我们的忆阻器将电阻变化降低了大约 30%，并且推理精度在类似范围内相应提高。精度增量通过我们的深度延迟反馈储层计算（Deep-DFR）模型进行评估。与传统静态随机存取存储器技术（6T）相比，设计面积、功耗和延迟分别减少了约48%、约42%和约67%。与最先进的忆阻器相比，我们的忆阻器的性能有不同程度的提高（大约13%-73%）。
Deep neural networks (DNNs), a brain-inspired learning methodology, requires tremendous data for training before performing inference tasks. The recent studies demonstrate a strong positive correlation between the inference accuracy and the size of the DNNs and datasets, which leads to an inevitable demand for large DNNs. However, conventional memory techniques are not adequate to deal with the drastic growth of dataset and neural network size. Recently, a resistive memristor has been widely considered as the next generation memory device owing to its high density and low power consumption. Nevertheless, its high switching resistance variations (cycle-to-cycle) restrict its feasibility in deep learning. In this work, a novel memristor configuration with the enhanced heat dissipation feature is fabricated and evaluated to address this challenge. Our experimental results demonstrate our memristor reduces the resistance variation by similar to 30% and the inference accuracy increases correspondingly in a similar range. The accuracy increment is evaluated by our deep delay-feed-back reservoir computing (Deep-DFR) model. The design area, power consumption, and latency are reduced by similar to 48%, similar to 42%, and similar to 67%, respectively, compared to the conventional static random-access memory technique (6T). The performance of our memristor is improved at various degrees (similar to 13%-73%) compared to the state-of-the-art memristors.