Spike-Based Plasticity Circuits for Always-on On-Line Learning in Neuromorphic Systems

Event-driven neuromorphic hardware with on-line learning capabilities enables the low-power local processing of signals on the edge sensors. Implementing such hardware requires having an always-on online learning operation in order to continuously adapt to the changes in the environment. Therefore, as the data is continuously streaming, there cannot be a separation between the training and the testing phase. Such constraint thus asks for a continuous time learning strategy which includes a mechanism to stop changing the weights when the system has reached an optimal operating point, so that it does not over-fit the input data and it generalizes to unseen patterns of the learned class. In this paper we propose spike-based circuits based on a local gradient-descent based learning rule that comprise also this additional “stop-learning” feature and that have a wide range of configurability options over the learning parameters. We describe the circuit behavior and present simulation results for a standard CMOS 180 nm process, showing how the width of the stop-learning region can be controlled along with the learning rate of the system. Such system represents a hardware implementation of a feature which has shown to improves the stability of the learning process and the convergence properties of the network.

[1]  Farnood Merrikh-Bayat,et al.  Self-Adaptive Spike-Time-Dependent Plasticity of Metal-Oxide Memristors , 2015, Scientific Reports.

[2]  Somnath Paul,et al.  Event-Driven Random Back-Propagation: Enabling Neuromorphic Deep Learning Machines , 2016, Front. Neurosci..

[3]  Yoshua Bengio,et al.  Dendritic error backpropagation in deep cortical microcircuits , 2017, ArXiv.

[4]  Richard George,et al.  Spike-driven threshold-based learning with memristive synapses and neuromorphic silicon neurons , 2018, Journal of Physics D: Applied Physics.

[5]  Giacomo Indiveri,et al.  A neuromorphic systems approach to in-memory computing with non-ideal memristive devices: From mitigation to exploitation , 2018, Faraday discussions.

[6]  Walter Senn,et al.  Learning Real-World Stimuli in a Neural Network with Spike-Driven Synaptic Dynamics , 2007, Neural Computation.

[7]  Bradley A. Minch.  A simple variable-width CMOS bump circuit , 2016, 2016 IEEE International Symposium on Circuits and Systems (ISCAS).

[8]  Anders Krogh,et al.  Introduction to the theory of neural computation , 1994, The advanced book program.

[9]  Kazuyuki Aihara,et al.  A CMOS Spiking Neural Network Circuit with Symmetric/Asymmetric STDP Function , 2009, IEICE Trans. Fundam. Electron. Commun. Comput. Sci..

[10]  Giacomo Indiveri,et al.  A differential memristive synapse circuit for on-line learning in neuromorphic computing systems , 2017, ArXiv.

[11]  Giacomo Indiveri,et al.  Event-based circuits for controlling stochastic learning with memristive devices in neuromorphic architectures , 2018, 2018 IEEE International Symposium on Circuits and Systems (ISCAS).

[12]  Giacomo Indiveri,et al.  On-chip unsupervised learning in winner-take-all networks of spiking neurons , 2017, 2017 IEEE Biomedical Circuits and Systems Conference (BioCAS).

[13]  Giacomo Indiveri,et al.  A reconfigurable on-line learning spiking neuromorphic processor comprising 256 neurons and 128K synapses , 2015, Front. Neurosci..

[14]  Emre O. Neftci,et al.  Data and Power Efficient Intelligence with Neuromorphic Learning Machines , 2018, iScience.

[15]  David Bol,et al.  A 0.086-mm$^2$ 12.7-pJ/SOP 64k-Synapse 256-Neuron Online-Learning Digital Spiking Neuromorphic Processor in 28-nm CMOS , 2018, IEEE Transactions on Biomedical Circuits and Systems.

[16]  Carver A. Mead.  Analog VLSI and neural systems (invited presentation) , 1989 .

[17]  Carlo Baldassi,et al.  Learning may need only a few bits of synaptic precision. , 2016, Physical review. E.

[18]  T. Delbruck.  'Bump' circuits for computing similarity and dissimilarity of analog voltages , 1991, IJCNN-91-Seattle International Joint Conference on Neural Networks.

[19]  Bernabé Linares-Barranco,et al.  On Spike-Timing-Dependent-Plasticity, Memristive Devices, and Building a Self-Learning Visual Cortex , 2011, Front. Neurosci..

[20]  Johannes Schemmel,et al.  Modeling Synaptic Plasticity within Networks of Highly Accelerated I&F Neurons , 2007, 2007 IEEE International Symposium on Circuits and Systems.

[21]  Melika Payvand,et al.  A CMOS-memristive self-learning neural network for pattern classification applications , 2014, 2014 IEEE/ACM International Symposium on Nanoscale Architectures (NANOARCH).

[22]  Chiara Bartolozzi,et al.  Neuromorphic Electronic Circuits for Building Autonomous Cognitive Systems , 2014, Proceedings of the IEEE.