Resistive random-access memory (ReRAM) is a promising technology for designing hardware accelerators for deep neural network (DNN) inferencing. However, stochastic noise in ReRAM crossbars can degrade the DNN inferencing accuracy. We propose the design and optimization of a high-performance, area-and energy-efficient ReRAM-based hardware accelerator to achieve robust DNN inferencing in the presence of stochastic noise. We make two key technical contributions. First, we propose a stochastic-noise-aware training method, referred to as ReSNA, to improve the accuracy of DNN inferencing on ReRAM crossbars with stochastic noise. Second, we propose an information-theoretic algorithm, referred to as CF-MESMO, to identify the Pareto set of solutions to trade-off multiple objectives, including inferencing accuracy, area overhead, execution time, and energy consumption. The main challenge in this context is that executing the ReSNA method to evaluate each candidate ReRAM design is prohibitive. To address this challenge, we utilize the continuous-fidelity evaluation of ReRAM designs associated with prohibitive high computation cost by varying the number of training epochs to trade-off accuracy and cost. CF-MESMO iteratively selects the candidate ReRAM design and fidelity pair that maximizes the information gained per unit computation cost about the optimal Pareto front. Our experiments on benchmark DNNs show that the proposed algorithms efficiently uncover high-quality Pareto fronts. On average, ReSNA achieves 2.57% inferencing accuracy improvement for ResNet20 on the CIFAR-10 dataset with respect to the baseline configuration. Moreover, CF-MESMO algorithm achieves 90.91% reduction in computation cost compared to the popular multi-objective optimization algorithm NSGA-II to reach the best solution from NSGA-II.

中文翻译：

---

随机噪声下鲁棒 DNN 推理的 ReRAM Crossbars 多目标优化

电阻式随机存取存储器 (ReRAM) 是一种很有前途的技术，用于设计用于深度神经网络 (DNN) 推理的硬件加速器。然而，ReRAM crossbar 中的随机噪声会降低 DNN 推理的准确性。我们建议设计和优化高性能、面积和节能的基于 ReRAM 的硬件加速器，以在存在随机噪声的情况下实现稳健的 DNN 推理。我们做出了两个关键的技术贡献。首先，我们提出了一种随机噪声感知训练方法，称为 ReSNA，以提高 DNN 对具有随机噪声的 ReRAM crossbar 进行推理的准确性。其次，我们提出了一种称为 CF-MESMO 的信息论算法，以识别帕累托解决方案集以权衡多个目标，包括推理精度、面积开销、执行时间、和能源消耗。这种情况下的主要挑战是执行 ReSNA 方法来评估每个候选 ReRAM 设计是令人望而却步的。为了应对这一挑战，我们通过改变训练时期的数量来权衡准确性和成本，从而利用与高昂的计算成本相关的 ReRAM 设计的连续保真度评估。CF-MESMO 迭代地选择候选 ReRAM 设计和保真度对，以最大化每单位计算成本获得的关于最佳帕累托前沿的信息。我们在基准 DNN 上的实验表明，所提出的算法有效地揭示了高质量的帕累托前沿。平均而言，相对于基线配置，ReSNA 在 CIFAR-10 数据集上对 ResNet20 的推理精度提高了 2.57%。此外，CF-MESMO算法达到了90。