An approximate backpropagation learning rule for memristor based neural networks using synaptic plasticity

We describe an approximation to backpropagation algorithm for training deep neural networks, which is designed to work with synapses implemented with memristors. The key idea is to represent the values of both the input signal and the backpropagated delta value with a series of pulses that trigger multiple positive or negative updates of the synaptic weight, and to use the min operation instead of the product of the two signals. In computational simulations, we show that the proposed approximation to backpropagation is well converged and may be suitable for memristor implementations of multilayer neural networks.

[1]  G. G. Stokes.  "J." , 1890, The New Yale Book of Quotations.

[2]  Leon O. Chua,et al.  Memristor Bridge Synapses , 2012, Proceedings of the IEEE.

[3]  Jacques-Olivier Klein,et al.  Robust neural logic block (NLB) based on memristor crossbar array , 2011, 2011 IEEE/ACM International Symposium on Nanoscale Architectures.

[4]  Daniel Cownden,et al.  Random feedback weights support learning in deep neural networks , 2014, ArXiv.

[5]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

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

[7]  Leon O. Chua,et al.  Everything You Wish to Know About Memristors but Are Afraid to Ask , 2015, Handbook of Memristor Networks.

[8]  Huamin Wang,et al.  Pavlov associative memory in a memristive neural network and its circuit implementation , 2016, Neurocomputing.

[9]  Tarek M. Taha,et al.  Enabling back propagation training of memristor crossbar neuromorphic processors , 2014, 2014 International Joint Conference on Neural Networks (IJCNN).

[10]  Luca Maria Gambardella,et al.  Better Digit Recognition with a Committee of Simple Neural Nets , 2011, 2011 International Conference on Document Analysis and Recognition.

[11]  O. Richard,et al.  10×10nm2 Hf/HfOx crossbar resistive RAM with excellent performance, reliability and low-energy operation , 2011, 2011 International Electron Devices Meeting.

[12]  Giacomo Indiveri,et al.  Integration of nanoscale memristor synapses in neuromorphic computing architectures , 2013, Nanotechnology.

[14]  Bernabé Linares-Barranco,et al.  On neuromorphic spiking architectures for asynchronous STDP memristive systems , 2010, Proceedings of 2010 IEEE International Symposium on Circuits and Systems.

[15]  Mika Laiho,et al.  Cellular nanoscale network cell with memristors for local implication logic and synapses , 2010, Proceedings of 2010 IEEE International Symposium on Circuits and Systems.

[16]  Garrett S. Rose,et al.  Memristor-Based Neural Logic Blocks for Nonlinearly Separable Functions , 2013, IEEE Transactions on Computers.

[17]  Bernard Widrow,et al.  The No-Prop algorithm: A new learning algorithm for multilayer neural networks , 2013, Neural Networks.

[18]  Gert Cauwenberghs,et al.  Neuromorphic Silicon Neuron Circuits , 2011, Front. Neurosci.

[19]  A. Ayatollahi,et al.  Implementation of biologically plausible spiking neural network models on the memristor crossbar-based CMOS/nano circuits , 2009, 2009 European Conference on Circuit Theory and Design.

[20]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

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

[22]  Piotr Dudek,et al.  Gradient-descent-based learning in memristive crossbar arrays , 2015, 2015 International Joint Conference on Neural Networks (IJCNN).

[23]  Shukai Duan,et al.  Multilayer RTD-memristor-based cellular neural networks for color image processing , 2015, Neurocomputing.

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

[25]  Leon O. Chua,et al.  A Circuit-Based Learning Architecture for Multilayer Neural Networks With Memristor Bridge Synapses , 2015, IEEE Transactions on Circuits and Systems I: Regular Papers.

[26]  Jawar Singh,et al.  Linearly separable pattern classification using memristive crossbar circuits , 2014, Fifteenth International Symposium on Quality Electronic Design.

[27]  Cory Merkel,et al.  Heterogeneous CMOS/memristor hardware neural networks for real-time target classification , 2014, Sensing Technologies + Applications.

[28]  Konstantin K. Likharev,et al.  CrossNets: Neuromorphic Hybrid CMOS/Nanoelectronic Networks , 2011 .

[29]  P. Werbos,et al.  Beyond Regression : "New Tools for Prediction and Analysis in the Behavioral Sciences , 1974 .

[30]  Damien Querlioz,et al.  Simulation of a memristor-based spiking neural network immune to device variations , 2011, The 2011 International Joint Conference on Neural Networks.

[31]  A. Zenkevich,et al.  Multilevel resistive switching in ternary HfxAl1-xOy oxide with graded Al depth profile , 2013 .

[32]  Jason Cong,et al.  Optimizing FPGA-based Accelerator Design for Deep Convolutional Neural Networks , 2015, FPGA.

[33]  Martin A. Riedmiller,et al.  A direct adaptive method for faster backpropagation learning: the RPROP algorithm , 1993, IEEE International Conference on Neural Networks.

[34]  A. Zenkevich,et al.  Resistive switching and synaptic properties of fully atomic layer deposition grown TiN/HfO2/TiN devices , 2015 .

[35]  Dhireesha Kudithipudi,et al.  Comparison of Off-Chip Training Methods for Neuromemristive Systems , 2015, 2015 28th International Conference on VLSI Design.

[36]  Avinoam Kolodny,et al.  Memristor-Based Multilayer Neural Networks With Online Gradient Descent Training , 2015, IEEE Transactions on Neural Networks and Learning Systems.

[37]  Byoungil Lee,et al.  Nanoelectronic programmable synapses based on phase change materials for brain-inspired computing. , 2012, Nano letters.

[38]  Pinaki Mazumder,et al.  CMOS and Memristor-Based Neural Network Design for Position Detection , 2012, Proceedings of the IEEE.

[39]  Leon O. Chua,et al.  Neural Synaptic Weighting With a Pulse-Based Memristor Circuit , 2012, IEEE Transactions on Circuits and Systems I: Regular Papers.