当前位置: X-MOL 学术IEEE Electron Device Lett. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
High precision symmetric weight update of memristor by gate voltage ramping method for convolutional neural network accelerator
IEEE Electron Device Letters ( IF 4.1 ) Pub Date : 2020-03-01 , DOI: 10.1109/led.2020.2968388
Jia Chen , Wen-Qian Pan , Yi Li , Rui Kuang , Yu-Hui He , Chih-Yang Lin , Nian Duan , Gui-Rong Feng , Hao-Xuan Zheng , Ting-Chang Chang , Simon M. Sze , Xiang-Shui Miao

Memristor emerges as the key enabler for neural network accelerator. Here, we demonstrate high-precision symmetric weight update in a one transistor one resistor (1T1R) structure Ti/HfO2/TiN memristor using a gate voltage ramping method, with over 120-level states and low variation (< 4%). Incorporating all experimental non-idealities, the proposed mixed hardware-software convolutional neural network demonstrates over 92.79% online learning accuracy (against software equivalent 98.45%) for MNIST recognition task. The network also shows robustness to input image noises, array yield, and retention issues.

中文翻译:

卷积神经网络加速器栅极电压斜坡法高精度对称权重更新

忆阻器成为神经网络加速器的关键推动者。在这里,我们展示了使用栅极电压斜坡方法在单晶体管单电阻 (1T1R) 结构 Ti/HfO2/TiN 忆阻器中的高精度对称权重更新,具有超过 120 级状态和低变化 (< 4%)。结合所有实验性的非理想性,所提出的混合硬件 - 软件卷积神经网络在 MNIST 识别任务中展示了超过 92.79% 的在线学习准确率(相对于软件等效 98.45%)。该网络还显示出对输入图像噪声、阵列良率和保留问题的鲁棒性。
更新日期:2020-03-01
down
wechat
bug