Voltage-dependent synaptic plasticity: Unsupervised probabilistic Hebbian plasticity rule based on neurons membrane potential

This study proposes voltage-dependent-synaptic plasticity (VDSP), a novel brain-inspired unsupervised local learning rule for the online implementation of Hebb’s plasticity mechanism on neuromorphic hardware. The proposed VDSP learning rule updates the synaptic conductance on the spike of the postsynaptic neuron only, which reduces by a factor of two the number of updates with respect to standard spike timing dependent plasticity (STDP). This update is dependent on the membrane potential of the presynaptic neuron, which is readily available as part of neuron implementation and hence does not require additional memory for storage. Moreover, the update is also regularized on synaptic weight and prevents explosion or vanishing of weights on repeated stimulation. Rigorous mathematical analysis is performed to draw an equivalence between VDSP and STDP. To validate the system-level performance of VDSP, we train a single-layer spiking neural network (SNN) for the recognition of handwritten digits. We report 85.01 ± 0.76% (Mean ± SD) accuracy for a network of 100 output neurons on the MNIST dataset. The performance improves when scaling the network size (89.93 ± 0.41% for 400 output neurons, 90.56 ± 0.27 for 500 neurons), which validates the applicability of the proposed learning rule for spatial pattern recognition tasks. Future work will consider more complicated tasks. Interestingly, the learning rule better adapts than STDP to the frequency of input signal and does not require hand-tuning of hyperparameters.

[1]  V. A. Demin,et al.  Necessary conditions for STDP-based pattern recognition learning in a memristive spiking neural network , 2020, Neural Networks.

[2]  Jordi Madrenas,et al.  Analog-circuit implementation of multiplicative spike-timing-dependent plasticity with linear decay , 2021, Nonlinear Theory and Its Applications, IEICE.

[3]  Ruholla Jafari-Marandi.  Supervised or unsupervised learning? Investigating the role of pattern recognition assumptions in the success of binary predictive prescriptions , 2021, Neurocomputing.

[4]  Mauricio Barahona,et al.  Learning compositional sequences with multiple time scales through a hierarchical network of spiking neurons , 2020, bioRxiv.

[5]  Bernabé Linares-Barranco,et al.  Implementation of a tunable spiking neuron for STDP with memristors in FDSOI 28nm , 2020, 2020 2nd IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS).

[6]  Binsu J Kailath,et al.  A Novel Method to Implement STDP Learning Rule in Verilog , 2020, 2020 IEEE Region 10 Symposium (TENSYMP).

[7]  Andreas Grübl,et al.  Verification and Design Methods for the BrainScaleS Neuromorphic Hardware System , 2020, Journal of Signal Processing Systems.

[8]  Kirk Y. W. Scheper,et al.  Unsupervised Learning of a Hierarchical Spiking Neural Network for Optical Flow Estimation: From Events to Global Motion Perception , 2018, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[9]  Mikhail Zymbler,et al.  Internet of Things is a revolutionary approach for future technology enhancement: a review , 2019, Journal of Big Data.

[10]  Hesham Mostafa,et al.  Surrogate Gradient Learning in Spiking Neural Networks: Bringing the Power of Gradient-based optimization to spiking neural networks , 2019, IEEE Signal Processing Magazine.

[11]  Jong-Ho Lee,et al.  Unsupervised online learning of temporal information in spiking neural network using TFT-type NOR flash memory devices. , 2019, Nanotechnology.

[12]  He Qian,et al.  Unsupervised Learning on Resistive Memory Array Based Spiking Neural Networks , 2019, Front. Neurosci..

[13]  Tara Julia Hamilton,et al.  Efficient FPGA Implementations of Pair and Triplet-Based STDP for Neuromorphic Architectures , 2019, IEEE Transactions on Circuits and Systems I: Regular Papers.

[14]  Peter Blouw,et al.  Benchmarking Keyword Spotting Efficiency on Neuromorphic Hardware , 2018, NICE '19.

[15]  Gopalakrishnan Srinivasan,et al.  Training Deep Spiking Convolutional Neural Networks With STDP-Based Unsupervised Pre-training Followed by Supervised Fine-Tuning , 2018, Front. Neurosci..

[16]  Hermann Kohlstedt,et al.  A memristive plasticity model of voltage-based STDP suitable for recurrent bidirectional neural networks in the hippocampus , 2018, Scientific Reports.

[17]  Yusuf Leblebici,et al.  Neuromorphic computing with multi-memristive synapses , 2017, Nature Communications.

[18]  Timothée Masquelier,et al.  STDP-based spiking deep neural networks for object recognition , 2016, Neural Networks.

[19]  S. Thorpe,et al.  STDP-based spiking deep convolutional neural networks for object recognition , 2018 .

[20]  Bernabé Linares-Barranco,et al.  Hardware implementation of convolutional STDP for on-line visual feature learning , 2017, 2017 IEEE International Symposium on Circuits and Systems (ISCAS).

[21]  Johannes Schemmel,et al.  Demonstrating Hybrid Learning in a Flexible Neuromorphic Hardware System , 2016, IEEE Transactions on Biomedical Circuits and Systems.

[22]  Terrence J. Sejnowski,et al.  Unsupervised Learning , 2018, Encyclopedia of GIS.

[23]  Chip-Hong Chang,et al.  A low-voltage, low power STDP synapse implementation using domain-wall magnets for spiking neural networks , 2016, 2016 IEEE International Symposium on Circuits and Systems (ISCAS).

[24]  Paolo Fantini,et al.  Unsupervised Learning by Spike Timing Dependent Plasticity in Phase Change Memory (PCM) Synapses , 2016, Front. Neurosci..

[25]  Lubica Benusková,et al.  A Voltage-Based STDP Rule Combined with Fast BCM-Like Metaplasticity Accounts for LTP and Concurrent “Heterosynaptic” LTD in the Dentate Gyrus In Vivo , 2015, PLoS Comput. Biol..

[26]  Matthew Cook,et al.  Unsupervised learning of digit recognition using spike-timing-dependent plasticity , 2015, Front. Comput. Neurosci..

[27]  Matthew Cook,et al.  Efficient implementation of STDP rules on SpiNNaker neuromorphic hardware , 2014, 2014 International Joint Conference on Neural Networks (IJCNN).

[28]  Tomoki Fukai,et al.  Computational Implications of Lognormally Distributed Synaptic Weights , 2014, Proceedings of the IEEE.

[29]  Trevor Bekolay,et al.  Nengo: a Python tool for building large-scale functional brain models , 2014, Front. Neuroinform..

[30]  D. Querlioz,et al.  Immunity to Device Variations in a Spiking Neural Network With Memristive Nanodevices , 2013, IEEE Transactions on Nanotechnology.

[31]  T. Serrano-Gotarredona,et al.  STDP and STDP variations with memristors for spiking neuromorphic learning systems , 2013, Front. Neurosci..

[32]  Jacques-Olivier Klein,et al.  Bioinspired networks with nanoscale memristive devices that combine the unsupervised and supervised learning approaches , 2012, 2012 IEEE/ACM International Symposium on Nanoscale Architectures (NANOARCH).

[33]  Matthieu Gilson,et al.  Frontiers in Computational Neuroscience Computational Neuroscience , 2022 .

[34]  W. Gerstner,et al.  Connectivity reflects coding: a model of voltage-based STDP with homeostasis , 2010, Nature Neuroscience.

[35]  W. Gerstner,et al.  Connectivity reflects coding: A model of voltage-based spike-timing-dependent-plasticity with homeostasis , 2009 .

[36]  Wulfram Gerstner,et al.  Phenomenological models of synaptic plasticity based on spike timing , 2008, Biological Cybernetics.

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

[38]  Timothée Masquelier,et al.  Unsupervised Learning of Visual Features through Spike Timing Dependent Plasticity , 2007, PLoS Comput. Biol..

[39]  Nitesh V. Chawla,et al.  Learning From Labeled And Unlabeled Data: An Empirical Study Across Techniques And Domains , 2011, J. Artif. Intell. Res..

[40]  Walter Senn,et al.  Minimal Models of Adapted Neuronal Response to In VivoLike Input Currents , 2004, Neural Computation.

[41]  Mark C. W. van Rossum,et al.  Stable Hebbian Learning from Spike Timing-Dependent Plasticity , 2000, The Journal of Neuroscience.

[42]  L. Abbott,et al.  Competitive Hebbian learning through spike-timing-dependent synaptic plasticity , 2000, Nature Neuroscience.

[43]  L. F Abbott,et al.  Lapicque’s introduction of the integrate-and-fire model neuron (1907) , 1999, Brain Research Bulletin.

[44]  G. Bi,et al.  Synaptic Modifications in Cultured Hippocampal Neurons: Dependence on Spike Timing, Synaptic Strength, and Postsynaptic Cell Type , 1998, The Journal of Neuroscience.

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

[46]  V. Han,et al.  Synaptic plasticity in a cerebellum-like structure depends on temporal order , 1997, Nature.

[47]  Wolfgang Maass,et al.  Networks of Spiking Neurons: The Third Generation of Neural Network Models , 1996, Electron. Colloquium Comput. Complex..

[48]  T. Bliss,et al.  A synaptic model of memory: long-term potentiation in the hippocampus , 1993, Nature.

[49]  W. Singer,et al.  Different voltage-dependent thresholds for inducing long-term depression and long-term potentiation in slices of rat visual cortex , 1990, Nature.

[50]  J. Knott.  The organization of behavior: A neuropsychological theory , 1951 .