神经网络到Crossbar阵列的设备非理想弹性映射方法
A Device Non-Ideality Resilient Approach for Mapping Neural Networks to Crossbar Arrays
摘要：
我们提出了一种技术独立的方法，称为相邻连接矩阵（ACM），有效地映射到非负交叉阵列的符号权重矩阵。与相同硬件开销的映射方法相比，当使用5位精度或更低的crossbar阵列进行训练时，使用ACM可以将ResNet-20与CIFAR-10数据集的训练精度提高高达20%。与使用两个元素来表示权重的策略相比，ACM实现了相当的训练精度，同时还分别提供了2.3倍和7倍的面积和读取能量减少。ACM还具有温和的正则化效应，可以提高交叉阵列中的推理精度，而无需任何重新训练或昂贵的设备/变化感知训练。
英文摘要：
We propose a technology-independent method, referred to as adjacent connection matrix (ACM), to efficiently map signed weight matrices to non-negative crossbar arrays. When compared to same-hardware-overhead mapping methods, using ACM leads to improvements of up to 20% in training accuracy for ResNet-20 with the CIFAR-10 dataset when training with 5-bit precision crossbar arrays or lower. When compared with strategies that use two elements to represent a weight, ACM achieves comparable training accuracies, while also offering area and read energy reductions of 2.3x and 7x, respectively. ACM also has a mild regularization effect that improves inference accuracy in crossbar arrays without any retraining or costly device/variation-aware training.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
