Vortex: Variation-aware training for memristor X-bar

Recent advances in development of memristor devices and cross-bar integration allow us to implement a low-power on-chip neuromorphic computing system (NCS) with small footprint. Training methods have been proposed to program the memristors in a crossbar by following existing training algorithms in neural network models. However, the robustness of these training methods has not been well investigated by taking into account the limits imposed by realistic hardware implementations. In this work, we present a quantitative analysis on the impact of device imperfections and circuit design constraints on the robustness of two popular training methods - “close-loop on-device” (CLD) and “open-loop off-device” (OLD). A novel variation-aware training scheme, namely, Vortex, is then invented to enhance the training robustness of memristor crossbar-based NCS by actively compensating the impact of device variations and optimizing the mapping scheme from computations to crossbars. On average, Vortex can significantly improve the test rate by 29.6% and 26.4%, compared to the traditional OLD and CLD, respectively.

[1]  L. Chua Memristor-The missing circuit element , 1971 .

[2]  Wei Zhang,et al.  Digital-assisted noise-eliminating training for memristor crossbar-based analog neuromorphic computing engine , 2013, 2013 50th ACM/EDAC/IEEE Design Automation Conference (DAC).

[3]  Yiran Chen,et al.  BSB training scheme implementation on memristor-based circuit , 2013, 2013 IEEE Symposium on Computational Intelligence for Security and Defense Applications (CISDA).

[4]  Yiran Chen,et al.  Memristor crossbar based hardware realization of BSB recall function , 2012, The 2012 International Joint Conference on Neural Networks (IJCNN).

[5]  Fabien Alibart,et al.  Pattern classification by memristive crossbar circuits using ex situ and in situ training , 2013, Nature Communications.

[6]  U-In Chung,et al.  Multi-level switching of triple-layered TaOx RRAM with excellent reliability for storage class memory , 2012, 2012 Symposium on VLSI Technology (VLSIT).

[7]  Vibhore Vardhan,et al.  Power Consumption Breakdown on a Modern Laptop , 2004, PACS.

[8]  Qing Wu,et al.  Hardware realization of BSB recall function using memristor crossbar arrays , 2012, DAC Design Automation Conference 2012.

[9]  Sally A. McKee,et al.  Reflections on the memory wall , 2004, CF '04.

[10]  Rodney X. Sturdivant,et al.  Applied Logistic Regression: Hosmer/Applied Logistic Regression , 2005 .

[11]  D. Stewart,et al.  The missing memristor found , 2008, Nature.

[12]  Jiale Liang,et al.  Cross-Point Memory Array Without Cell Selectors—Device Characteristics and Data Storage Pattern Dependencies , 2010, IEEE Transactions on Electron Devices.

[13]  David W. Hosmer,et al.  Applied Logistic Regression , 1991 .

[14]  Kaushik Roy,et al.  Ultra low power associative computing with spin neurons and resistive crossbar memory , 2013, 2013 50th ACM/EDAC/IEEE Design Automation Conference (DAC).

[15]  Marc'Aurelio Ranzato,et al.  Building high-level features using large scale unsupervised learning , 2011, 2013 IEEE International Conference on Acoustics, Speech and Signal Processing.

[16]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[17]  Shimeng Yu,et al.  Investigating the switching dynamics and multilevel capability of bipolar metal oxide resistive switching memory , 2011 .

[18]  Yiran Chen,et al.  Reduction and IR-drop compensations techniques for reliable neuromorphic computing systems , 2014, 2014 IEEE/ACM International Conference on Computer-Aided Design (ICCAD).