PAWN: Programmed Analog Weights for Non-Linearity Optimization in Memristor-Based Neuromorphic Computing System
PAWN: Programmed Analog Weights for Non-Linearity Optimization in Memristor-Based Neuromorphic Computing System
复制标题
DOI：
10.1109/jetcas.2023.3235658
复制
发表时间：
2023-03
影响因子：
4.6
通讯作者：
Saleh Ahmad Khan;Md. Oli-Uz-Zaman;Jinhui Wang
中科院分区：
文献类型：
--
作者：
Saleh Ahmad Khan;Md. Oli-Uz-Zaman;Jinhui Wang
忆阻器作为神经形态系统的硬件解决方案提供了优势。然而，它们的非线性器件特性使得权值更新不准确，降低了神经网络的推理精度。本文提出了一种非线性编程模拟权重（PAWN）方法，用于在神经形态系统的训练过程中，通过跟踪非线性曲线来更新忆阻器的电导。实验结果表明，PAWN方法在不同的LTP/LTD条件下都能有效地抑制非线性对忆阻器的影响。特别是在极端非线性（LTP=6，LTD=-6）下，基于忆阻器的神经形态系统的准确度非常低（51.77%），但PAWN方法可以在不增加推理能量和延迟开销的情况下大幅提高准确度（9.87%）。此外，神经形态系统的整体性能也进行了评估，以进一步验证。最后，综合实验表明，PAWN方法仍然是非常有效的，即使考虑到器件到器件的变化，周期到周期的变化，各种技术节点，和不同的架构。
Memristors offer advantages as a hardware solution for neuromorphic systems. However, their nonlinear device property makes the weight update inaccurately and reduces the inference accuracy of a neural network. A Programmed Analog Weights for Nonlinearity (PAWN) method is proposed in this paper to update the conductance of a memristor by following the nonlinear curve during the training in a neuromorphic system. The experiment results indicates the PAWN method is effective to alleviate the nonlinearity influence to memristors in all different LTP/LTD conditions. Especially in extreme nonlinearity (LTP=6, LTD=−6), the memristor-based neuromorphic system has significantly low accuracy (51.77%), but the PAWN method enables large accuracy improvement (9.87%) without the inference energy and latency overhead. In addition, overall performance of the neuromorphic system is also evaluated for further verification. Finally, comprehensive experiments show that the PAWN method is still greatly valid even considering device-to-device variations, cycle-to-cycle variations, various technology nodes, and different architectures.