Filters








8,111 Hits in 9.6 sec

Efficient single input-output layer spiking neural classifier with time-varying weight model [article]

Abeegithan Jeyasothy, Savitha Ramasamy, Suresh Sundaram
2019 arXiv   pre-print
Thus, it can be inferred that a single input-output layer spiking neural network with time-varying weight model is computationally more efficient than a multi-layer spiking neural network with long-term  ...  This paper presents a supervised learning algorithm, namely, the Synaptic Efficacy Function with Meta-neuron based learning algorithm (SEF-M) for a spiking neural network with a time-varying weight model  ...  SEF-M rule has been developed for a single input-output layer spiking neural network classifier with the time-varying weight model.  ... 
arXiv:1904.10400v1 fatcat:ghrtrhxb55ebnbesvvwurzodiu

Neural Spike Sorting using Binarized Neural Networks

Daniel Valencia, Amir Alimohammad
2020 IEEE transactions on neural systems and rehabilitation engineering  
This article presents the design and efficient hardware implementation of binarized neural networks (BNNs) for brain-implantable neural spike sorting.  ...  To the best of our knowledge, this is the first work employing BNNs for real-time in vivo neural spike sorting.  ...  An ANN consists of an input layer, zero or more hidden layers, and one output layer with various number of ANs in each layer.  ... 
doi:10.1109/tnsre.2020.3043403 pmid:33296305 fatcat:3h67c5qsi5doph5acu5je6gbdq

Spiking Neural Network-based Structural Health Monitoring Hardware System

Aqib Javed, Jim Harkin, Liam McDaid, Junxiu Liu
2021 2021 IEEE Symposium Series on Computational Intelligence (SSCI)  
In this work, we propose an SNN based low-cost, energy-efficient, and standalone damage classification model for SHM.  ...  A key challenge is the ability to detect damages in an efficient manner for edge computing.  ...  A single spiking neuron at output layer can be trained to classify number of discrete states by varying spike firing time.  ... 
doi:10.1109/ssci50451.2021.9659844 fatcat:gthcxyvwvncapgp73fegt6k3hy

Unsupervised speech recognition through spike-timing-dependent plasticity in a convolutional spiking neural network

Meng Dong, Xuhui Huang, Bo Xu
2018 PLoS ONE  
The network adopts the time-to-first-spike coding scheme for fast and efficient information processing.  ...  These demonstrate that an STDP-based convolutional SNN model equipped with local weight sharing and temporal coding is capable of solving the SR task accurately and efficiently.  ...  We also gratefully acknowledge the support of NVIDIA Corporation with the donation of the Titan X Pascal GPU used for this research. Author Contributions Conceptualization: Meng Dong, Xuhui Huang.  ... 
doi:10.1371/journal.pone.0204596 pmid:30496179 pmcid:PMC6264808 fatcat:6ogx23t2qfdxnivhoouikdcxyy

Spike encoding for pattern recognition: Comparing cerebellum granular layer encoding and BSA algorithms

Chaitanya Medini, Ritu Maria Zacharia, Bipin Nair, Asha Vijayan, Lekshmi Priya Rajagopal, Shyam Diwakar
2015 2015 International Conference on Advances in Computing, Communications and Informatics (ICACCI)  
We developed a new spike encoding model inspired from cerebellum granular layer and tested different classification techniques like SVM, Naïve Bayes, MLP for training spiking neural networks to perform  ...  Spiking neural encoding models allow classification of real world tasks to suit for brain-machine interfaces in addition to serving as internal models.  ...  The neural network was created using simple AdEx [19] model which contain 2 layers, the first layer resembling the granular layer and the second layer resembling a single Purkinje neuron, whose output  ... 
doi:10.1109/icacci.2015.7275845 dblp:conf/icacci/MediniZNVRD15 fatcat:7efwxa5mnvc5hpkhatpruzizd4

A Dynamic Reconfigurable Architecture for Hybrid Spiking and Convolutional FPGA-Based Neural Network Designs

Hasan Irmak, Federico Corradi, Paul Detterer, Nikolaos Alachiotis, Daniel Ziener
2021 Journal of Low Power Electronics and Applications  
The second task requires switching between NN architectures, namely a CNN architecture with high accuracy and throughput and a hybrid architecture that combines convolutional layers and an optimized Spiking  ...  The first task is the image classification of two distinct datasets, and this requires switching between Convolutional Neural Network (CNN) architectures having different layer structures.  ...  In standard CNNs models, the parameters of each layer are the connection weights, the neuron biases, and the input and output of each layer are activations.  ... 
doi:10.3390/jlpea11030032 fatcat:pqlf64fxpnfz5agvagqkedd4vy

A basis coupled evolving spiking neural network with afferent input neurons

D. Shirin, R. Savitha, S. Suresh
2013 The 2013 International Joint Conference on Neural Networks (IJCNN)  
BCESNN is a two-layered neuron model with afferent neurons in the input layer and efferent neurons in the output layer.  ...  The afferent neurons in the input layer convert the real-valued input feature to a train of spikes using a bank of Gaussian Receptive Field (GRF) for each individual feature.  ...  BCESNN is a two-layered network with an input layer and an output layer.  ... 
doi:10.1109/ijcnn.2013.6706964 dblp:conf/ijcnn/ShirinSS13 fatcat:iyvbj55il5h5vglc337wghtnia

An Event-driven Recurrent Spiking Neural Network Architecture for Efficient Inference on FPGA

Anand Sankaran, Paul Detterer, Kalpana Kannan, Nikolaos Alachiotis, Federico Corradi
2022 Proceedings of the International Conference on Neuromorphic Systems 2022  
spike-time information and weight values.  ...  To this end, we propose a new digital architecture compatible with Recurrent Spiking Neural Networks (RSNNs) trained using the PyTorch framework and Back-Propagation-Through-Time (BPTT) for optimizing  ...  INTRODUCTION Spiking neural networks (SNNs) are efficient computational models that enable neural-inspired processing at the edge.  ... 
doi:10.1145/3546790.3546802 fatcat:hy25cjzfrvhnxj22uxo5izc7vm

Unsupervised clustering with spiking neurons by sparse temporal coding and multilayer RBF networks

S.M. Bohte, H. La Poutre, J.N. Kok
2002 IEEE Transactions on Neural Networks  
We demonstrate that spiking neural networks encoding information in spike times are capable of computing and learning clusters from realistic data.  ...  We develop a temporal encoding of continuously valued data to obtain adjustable clustering capacity and precision with an efficient use of neurons: input variables are encoded in a population code by neurons  ...  As a practical model, we use the Spike Response Model (SRM) introduced by Gerstner [9] , where the time-varying impact of a spike is described by a spike-response function.  ... 
doi:10.1109/72.991428 pmid:18244443 fatcat:aezwugdd2veblcxwjhsktloluu

Spike Pattern Transformation Learning In Structured Spiking Neural Networks

Brian Gardner, Andre Gruening
2015 Zenodo  
A preprint of our Neural Computation article (in press as of Sept 2015) can be found at http://arxiv.org/abs/1503.09129  ...  rules derivable using this method, and in particular their potential as highly efficient neural classifiers.  ...  The network's hidden layer varied in size, containing between 5 and 40 hidden neurons. For completeness, results for a single-layer structure with the same input layer size were included.  ... 
doi:10.5281/zenodo.30537 fatcat:utmeorrofngaplioodevuxuspi

A Multi-Spiking Neural Network Learning Model for Data Classification

Baagyere Edward Yellakuor, Agebure Apambila Moses, Qin Zhen, Oyetunji Elkanah Olaosebikan, Zhiguang Qin
2020 IEEE Access  
In order to address some of the inherent challenges associated with SNN, a multi-layer learning model for a multi-spiking network is proposed in this paper.  ...  It also employs a spike locality concept in order to determine how the synaptic weights are to be adjusted at a particular spike time so as to minimize the learning interference, and thereby, increasing  ...  The proposed model is derived based on a concept that exploits the relationship between input and output spikes times and synaptic weights that exist in a neural model [20] .  ... 
doi:10.1109/access.2020.2985257 fatcat:uu3fkgpcejddzpemmvjfwmcq5u

Locally Connected Spiking Neural Networks for Unsupervised Feature Learning [article]

Daniel J. Saunders, Devdhar Patel, Hananel Hazan, Hava T. Siegelmann, Robert Kozma
2019 arXiv   pre-print
We introduce a method for learning image features by locally connected layers in SNNs using spike-timing-dependent plasticity (STDP) rule.  ...  In recent years, Spiking Neural Networks (SNNs) have demonstrated great successes in completing various Machine Learning tasks.  ...  Learning rule Our locally-connected SNN layers are trained with a simple spike-timing-dependent plasticity (STDP) rule combined with a weight normalization scheme.  ... 
arXiv:1904.06269v1 fatcat:k7lc7nzchrezlfwbzrhzjaxel4

Self-organization of multi-layer spiking neural networks [article]

Guruprasad Raghavan, Cong Lin, Matt Thomson
2020 arXiv   pre-print
Inspired by this strategy, we attempt to efficiently self-organize large neural networks with an arbitrary number of layers into a wide variety of architectures.  ...  We also demonstrate that emergent waves can self-organize spiking network architecture to perform unsupervised learning, and networks can be coupled with a linear classifier to perform classification on  ...  Output :Weights W (l) (t) & spiking outputs y (l) (t) for all layers l ≥ 1. for t = 1...N t in ∆t time-steps do for l = 1...N l in layers do H (l) v,θ ← LIF (l) (x (l) , ∆t) integrate input with LIF by  ... 
arXiv:2006.06902v1 fatcat:3527vglixfbxnjv3odmklh7mle

Scalable energy-efficient, low-latency implementations of trained spiking Deep Belief Networks on SpiNNaker

Evangelos Stromatias, Daniel Neil, Francesco Galluppi, Michael Pfeiffer, Shih-Chii Liu, Steve Furber
2015 2015 International Joint Conference on Neural Networks (IJCNN)  
Importantly, using a neurally-inspired architecture yields additional benefits: during network run-time on this task, the platform consumes only 0.3 W with classification latencies in the order of tens  ...  spike activity rate.  ...  The latency of the hidden layers also varies with the number of input spikes.  ... 
doi:10.1109/ijcnn.2015.7280625 dblp:conf/ijcnn/StromatiasNGPLF15 fatcat:wz7o2atkyvbchmnjrnpxhf776i

FPT-spike: a flexible precise-time-dependent single-spike neuromorphic computing architecture

Tao Liu, Gang Quan, Wujie Wen
2020 CCF Transactions on High Performance Computing  
Recently the brain-inspired spiking neural network (SNN) has been demonstrated as a promising solution for delivering more impressive computing and power efficiency.  ...  For SNNs, a large body of prior work were conducted on the spiking system design with a focus on using the spike firing rate (or rate-coded) for fulfilling the practical cognitive tasks.  ...  We hope our study can inspire and motivate more in-depth research on the time-based SNN for realistic applications in energy-constraint platforms.  ... 
doi:10.1007/s42514-020-00037-6 fatcat:2hevpn5brfealictsg2vupzc6i
« Previous Showing results 1 — 15 out of 8,111 results