Wafer-Scale TaOx Device Variability and Implications for Neuromorphic Computing Applications

Scaling arrays of non-volatile memory devices from academic demonstrations to reliable, manufacturable systems requires a better understanding of variability at array and wafer-scale levels. CrossSim models the accuracy of neural networks implemented on an analog resistive memory accelerator using the cycle-to-cycle variability of a single device. In this work, we extend this modeling tool to account for device-to-device variation in a realistic way, and evaluate the impact of this reliability issue in the context of neuromorphic online learning tasks.

[1]  H. Hwang,et al.  Multi-layer tunnel barrier (Ta2O5/TaOx/TiO2) engineering for bipolar RRAM selector applications , 2013, Symposium on VLSI Technology.

[2]  Steven J. Plimpton,et al.  Resistive memory device requirements for a neural algorithm accelerator , 2016, 2016 International Joint Conference on Neural Networks (IJCNN).

[3]  Jonathan A. Cox,et al.  A Signal Processing Approach for Cyber Data Classification with Deep Neural Networks , 2015, Complex Adaptive Systems.

[4]  Steven J. Plimpton,et al.  Multiscale Co-Design Analysis of Energy, Latency, Area, and Accuracy of a ReRAM Analog Neural Training Accelerator , 2017, IEEE Journal on Emerging and Selected Topics in Circuits and Systems.

[5]  S. M. Dalton,et al.  Initial Assessment of the Effects of Radiation on the Electrical Characteristics of ${\rm TaO}_{\rm x}$ Memristive Memories , 2012, IEEE Transactions on Nuclear Science.

[6]  Qing Wu,et al.  Efficient and self-adaptive in-situ learning in multilayer memristor neural networks , 2018, Nature Communications.

[7]  Pritish Narayanan,et al.  Neuromorphic computing using non-volatile memory , 2017 .

[8]  Chao Du,et al.  Device nonideality effects on image reconstruction using memristor arrays , 2016, 2016 IEEE International Electron Devices Meeting (IEDM).

[9]  Kinam Kim,et al.  A fast, high-endurance and scalable non-volatile memory device made from asymmetric Ta2O(5-x)/TaO(2-x) bilayer structures. , 2011, Nature materials.

[10]  Jacques-Olivier Klein,et al.  Physical Realization of a Supervised Learning System Built with Organic Memristive Synapses , 2016, Scientific Reports.

[11]  Xiaochen Peng,et al.  NeuroSim: A Circuit-Level Macro Model for Benchmarking Neuro-Inspired Architectures in Online Learning , 2018, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.

[12]  S. E. Swanson,et al.  Training a Neural Network on Analog TaOx ReRAM Devices Irradiated With Heavy Ions: Effects on Classification Accuracy Demonstrated With CrossSim , 2019, IEEE Transactions on Nuclear Science.

[13]  Ru Huang,et al.  Modulation of nonlinear resistive switching behavior of a TaOx-based resistive device through interface engineering , 2017, Nanotechnology.