Spin-Hall MTJ Cells for Intra-Column Competition in Hierarchical Temporal Memory

We propose a dedicated winner-take-all circuit to efficiently implement the intra-column competition between cells in Hierarchical Temporal Memory which is a crucial part of the algorithm. All inputs and outputs are charge-based for compatibility with standard CMOS. The circuit incorporates memristors for competitive advantage to emulate a column with a cell in a predictive state. The circuit can also detect columns 'bursting' by passive averaging and comparison of the cell outputs. The proposed spintronic devices and circuit are thoroughly described and a series of simulations are used to predict the performance. The simulations indicate that the circuit can complete a nine-cell, nine-input competition operation in under 15 ns at a cost of about 25 pJ.

[1]  Kaushik Roy,et al.  Encoding Neural and Synaptic Functionalities in Electron Spin: A Pathway to Efficient Neuromorphic Computing , 2017, ArXiv.

[2]  Fabian Khateb,et al.  Sub 0.5-V bulk-driven winner take all circuit based on a new voltage follower , 2017 .

[3]  W. Rippard,et al.  Switching Distributions for Perpendicular Spin-Torque Devices Within the Macrospin Approximation , 2012, IEEE Transactions on Magnetics.

[4]  Jian-Gang Zhu,et al.  Magnetic tunnel junctions , 2006 .

[5]  Jennifer Hasler,et al.  Vector-Matrix Multiply and Winner-Take-All as an Analog Classifier , 2014, IEEE Transactions on Very Large Scale Integration (VLSI) Systems.

[6]  Yu-Cherng Hung,et al.  1-V bulk-driven CMOS analog programmable winner-takes-all circuit , 2006 .

[7]  Abdullah M. Zyarah,et al.  Neuromorphic Architecture for the Hierarchical Temporal Memory , 2019, IEEE Transactions on Emerging Topics in Computational Intelligence.

[8]  Shimeng Yu,et al.  Ferroelectric FET analog synapse for acceleration of deep neural network training , 2017, 2017 IEEE International Electron Devices Meeting (IEDM).

[9]  Jun Fu,et al.  A bio-inspired olfactory model using hierarchical temporal memory , 2012, 2012 5th International Conference on BioMedical Engineering and Informatics.

[10]  Youguang Zhang,et al.  Reconfigurable Codesign of STT-MRAM Under Process Variations in Deeply Scaled Technology , 2015, IEEE Transactions on Electron Devices.

[11]  Shukai Duan,et al.  An improved design of RBF neural network control algorithm based on spintronic memristor crossbar array , 2016, Neural Computing and Applications.

[12]  S. Natarajan,et al.  High sigma measurement of random threshold voltage variation in 14nm Logic FinFET technology , 2015, 2015 Symposium on VLSI Technology (VLSI Technology).

[13]  Azad Naeemi,et al.  Non-Boolean Computing Benchmarking for Beyond-CMOS Devices Based on Cellular Neural Network , 2016, IEEE Journal on Exploratory Solid-State Computational Devices and Circuits.

[14]  Wei Yang Lu,et al.  Nanoscale memristor device as synapse in neuromorphic systems. , 2010, Nano letters.

[15]  Lawrence T. Pileggi,et al.  mLogic: Ultra-low voltage non-volatile logic circuits using STT-MTJ devices , 2012, DAC Design Automation Conference 2012.

[16]  J. Hawkins,et al.  On Intelligence , 2004 .

[17]  D. Ralph,et al.  Spin transfer torques , 2007, 0711.4608.

[18]  D. Ralph,et al.  Spin transfer torque devices utilizing the giant spin Hall effect of tungsten , 2012, 1208.1711.

[19]  Zhaohao Wang,et al.  Spintronic logic design methodology based on spin Hall effect–driven magnetic tunnel junctions , 2016, Journal of Physics D: Applied Physics.

[20]  Abdullah M. Zyarah,et al.  Neuromemrisitive Architecture of HTM with On-Device Learning and Neurogenesis , 2019, ACM J. Emerg. Technol. Comput. Syst..

[21]  D. Ralph,et al.  Spin-Torque Switching with the Giant Spin Hall Effect of Tantalum , 2012, Science.

[22]  Supriyo Datta,et al.  Non-volatile spin switch for Boolean and non-Boolean logic , 2012 .

[23]  Wei Zhang,et al.  A thermal and process variation aware MTJ switching model and its applications in soft error analysis , 2012, 2012 IEEE/ACM International Conference on Computer-Aided Design (ICCAD).