Filters








7,803 Hits in 5.5 sec

Mapping Generative Models onto a Network of Digital Spiking Neurons

Bruno U. Pedroni, Srinjoy Das, John V. Arthur, Paul A. Merolla, Bryan L. Jackson, Dharmendra S. Modha, Kenneth Kreutz-Delgado, Gert Cauwenberghs
2016 IEEE Transactions on Biomedical Circuits and Systems  
Next, we describe the process of mapping generative RBMs trained offline onto the IBM TrueNorth neurosynaptic processor -- a low-power digital neuromorphic VLSI substrate.  ...  Generative performance metrics are analyzed to validate the neuromorphic requirements and to best select the neuron parameters for the model.  ...  ACKNOWLEDGEMENTS The authors would like to thank all the members of the Brain-Inspired Computing Group at the IBM Almaden Research Center for their dedication and collaboration.  ... 
doi:10.1109/tbcas.2016.2539352 pmid:27214915 fatcat:pfygt6j6tbc5rgpv2yuoyfgvbm

Neuromorphic architectures for spiking deep neural networks

Giacomo Indiveri, Federico Corradi, Ning Qiao
2015 2015 IEEE International Electron Devices Meeting (IEDM)  
We present a full custom hardware implementation of a deep neural network, built using multiple neuromorphic VLSI devices that integrate analog neuron and synapse circuits together with digital asynchronous  ...  The deep network comprises an event-based convolutional stage for feature extraction connected to a spike-based learning stage for feature classification.  ...  The spiking output of this network is then mapped onto the classification layer implemented on the ROLLS neuromorphic processor, which is configured to implement ensembles of perceptrons for classifying  ... 
doi:10.1109/iedm.2015.7409623 fatcat:4olpx74x55gv7i3edrcslowbze

Spiking Neural Networks Hardware Implementations and Challenges

Maxence Bouvier, Alexandre Valentian, Thomas Mesquida, Francois Rummens, Marina Reyboz, Elisa Vianello, Edith Beigne
2019 ACM Journal on Emerging Technologies in Computing Systems  
Recently, Spiking Neural Networks, a generation of cognitive algorithms employing computational primitives mimicking neuron and synapse operational principles, have become an important part of deep learning  ...  In this survey, we present the state of the art of hardware implementations of spiking neural networks and the current trends in algorithm elaboration from model selection to training mechanisms.  ...  a neural network mapped onto it [26, 76, 93] .  ... 
doi:10.1145/3304103 fatcat:p3frra3osnhybj4hkor4y5cyqm

Dynamical System Guided Mapping of Quantitative Neuronal Models Onto Neuromorphic Hardware

Peiran Gao, Ben V. Benjamin, Kwabena Boahen
2012 IEEE Transactions on Circuits and Systems Part 1: Regular Papers  
Hardware Behavior Neuron Behavior Model Analysis Linear Parametric Fit c Model Simulation Nonlinear Optimization No Model Behavior Matching b a  ...  ACKNOWLEDGMENT The authors would like to thank Emmett McQuinn for designing and programming the graphical user interface used for collecting spiking rates from mapped neurons.  ...  Mapping of only neuronal models is necessary but not sufficient for the construction of quantitatively accurate neural networks.  ... 
doi:10.1109/tcsi.2012.2188956 fatcat:wp33kjbwpzashlmejejvoopyze

Synthesizing cognition in neuromorphic electronic systems

E. Neftci, J. Binas, U. Rutishauser, E. Chicca, G. Indiveri, R. J. Douglas
2013 Proceedings of the National Academy of Sciences of the United States of America  
Here we report a solution to this problem that involves first mapping an unreliable hardware layer of spiking silicon neurons into an abstract computational layer composed of generic reliable subnetworks  ...  In the first step, the neural networks of the abstract layer are realized on the hardware substrate by mapping the neuron circuit bias voltages to the model parameters.  ...  Information and Communication Technologies Grant "acoustic SCene ANalysis for Detecting Living Entities (SCANDLE)" (231168), and by the Excellence Cluster 227 (Cognitive Interaction Technology-Center of  ... 
doi:10.1073/pnas.1212083110 pmid:23878215 pmcid:PMC3773754 fatcat:rup6b4bqjnh3lkccfyg5etwrnu

A Minimal Spiking Neural Network to Rapidly Train and Classify Handwritten Digits in Binary and 10-Digit Tasks

Amirhossein Tavanaei, Anthony S.
2015 International Journal of Advanced Research in Artificial Intelligence (IJARAI)  
The network architecture is primarily a feedforward spiking neural network (SNN) composed of Izhikevich regular spiking (RS) neurons and conductance-based synapses.  ...  The network uses biologically plausible neural and learning mechanisms and is applied to a subset of the MNIST dataset of handwritten digits.  ...  Therefore, the proposed SNN architecture and learning procedure can be a trustworthy model for classification due to its simple structure, quick feature extraction and learning, robust synaptic adaptation  ... 
doi:10.14569/ijarai.2015.040701 fatcat:a24vlq2o3bd2hk2vrobgzqorpm

A Digital Neuromorphic Architecture Efficiently Facilitating Complex Synaptic Response Functions Applied to Liquid State Machines [article]

Michael R. Smith, Aaron J. Hill, Kristofor D. Carlson, Craig M. Vineyard, Jonathon Donaldson, David R. Follett, Pamela L. Follett, John H. Naegle, Conrad D. James, James B. Aimone
2017 arXiv   pre-print
We consider the paradigm of spiking neurons with temporally coded information as opposed to non-spiking rate coded neurons used in most neural networks.  ...  Inspired by neuroscience principles, we present a digital neuromorphic architecture, the Spiking Temporal Processing Unit (STPU), capable of modeling arbitrary complex synaptic response functions without  ...  A high-level overview of a biological neuron and how its components map onto the STPU are shown in Figure 1 .  ... 
arXiv:1704.08306v1 fatcat:mvg6m7g5vnddtftd24kyawo56m

Artificial Cognitive Systems: From VLSI Networks of Spiking Neurons to Neuromorphic Cognition

Giacomo Indiveri, Elisabetta Chicca, Rodney J. Douglas
2009 Cognitive Computation  
Specifically, we show how VLSI networks of spiking neurons with spike-based plasticity mechanisms and soft winner-take-all architectures represent important building blocks useful for implementing artificial  ...  Neuromorphic engineering (NE) is an emerging research field that has been attempting to identify neural types of computational principles, by implementing biophysically realistic models of neural systems  ...  Acknowledgments This work was supported by the DAISY (FP6-2005-015803) EU Grant, by the Swiss National Science Foundation under Grant PMPD2-110298/1, and by the Swiss Federal Institute of Technology Zurich  ... 
doi:10.1007/s12559-008-9003-6 fatcat:gzrod52nxzgqzdifiedgqffwoi

Bio-Inspired Spiking Convolutional Neural Network using Layer-wise Sparse Coding and STDP Learning [article]

Amirhossein Tavanaei, Anthony S. Maida
2017 arXiv   pre-print
However, it is still not well understood how to create a biologically plausible network of brain-like, spiking neurons with multi-layer, unsupervised learning.  ...  The proposed model is evaluated on the MNIST digit dataset using clean and noisy images. The recognition performance for clean images is above 98%.  ...  Diehl et al (2015) reported 99.1% accuracy for their network converting an off-line trained CNN onto a spiking network [11] .  ... 
arXiv:1611.03000v4 fatcat:xvc7ofb4hvh23ffk5wgcor7dmi

Bio-inspired categorization using event-driven feature extraction and spike-based learning

Bo Zhao, Shoushun Chen, Huajin Tang
2014 2014 International Joint Conference on Neural Networks (IJCNN)  
The extracted spike feature patterns are then classified by a network of leaky integrate-and-fire (LIF) spiking neurons, in which the weights are trained using tempotron learning rule.  ...  One appealing characteristic of our system is the fully event-driven processing. The input, the features, and the classification are all based on address events (spikes).  ...  Each address event in the AER stream is projected onto a set of Gabor filters to build S1 feature maps. The pixel in S1 feature maps is modeled as a simplified LIF neuron.  ... 
doi:10.1109/ijcnn.2014.6889541 dblp:conf/ijcnn/ZhaoCT14 fatcat:damwclpqqrglbeaotjpk3tr36q

Massively distributed digital implementation of an integrate-and-fire LEGION network for visual scene segmentation

Bernard Girau, Cesar Torres-Huitzil
2007 Neurocomputing  
Results show that digital and flexible solutions may efficiently handle large networks of spiking neurons.  ...  Spiking neural networks offer an opportunity to develop models of visual perception without any complex structure based on multiple neural maps.  ...  As a result, a fully parallel implementation of the LEGION network has been mapped onto a Xilinx VIRTEX FPGA device, large enough to handle our low-resolution robot image sequences.  ... 
doi:10.1016/j.neucom.2006.11.009 fatcat:x5sjus3wirexrc7j4zwkmoyp6m

Large-scale neuromorphic computing systems

Steve Furber
2016 Journal of Neural Engineering  
We gratefully acknowledge all of these contributions.  ...  exploration of the capabilities of the machine is supported by the European Research Council under the European Union's Seventh Framework Programme (FP7/2007-2013)/ERC grant agreement 320689.  ...  Software running on the host will map the topology of the high-level model onto SpiNNaker's routing and compute resources and map the functionality of the model onto real-time library functions to be run  ... 
doi:10.1088/1741-2560/13/5/051001 pmid:27529195 fatcat:3ffwqgzpafetfgmac6m7ajkrqe

Neurogrid simulates cortical cell-types, active dendrites, and top-down attention [article]

Kwabena Boahen
2021 bioRxiv   pre-print
The resulting hybrid analog-digital platform, Neurogrid, scales to billions of synaptic connections, between up to a million neurons, and simulates cortical models in real-time using a few watts of electricity  ...  A central challenge for systems neuroscience and artificial intelligence is to understand how cognitive behaviors arise from large, highly interconnected networks of neurons.  ...  IEEE Transactions on Neural Networks 18, 1815-1825 (Nov. 2007). 21. Gao, P., Benjamin, V. & Boahen, K. Dynamical System Guided Mapping of Quantitative Neuronal Models Onto Neuromorphic Hardware.  ... 
doi:10.1101/2021.05.14.444265 fatcat:qvzfitbo7vghrgoephpb4ydrmm

System Design for In-Hardware STDP Learning and Spiking Based Probablistic Inference

Khadeer Ahmed, Amar Shrestha, Yanzhi Wang, Qinru Qiu
2016 2016 IEEE Computer Society Annual Symposium on VLSI (ISVLSI)  
A highly scalable and flexible digital hardware implementation of the neuron model is also presented.  ...  In this work, we present a set of neuron models and neuron circuit motifs that form SNNs capable of in-hardware fully-distributed STDP learning and spiking based probabilistic inference.  ...  Spiking neural networks (SNNs), which use spikes as the basis for communication, is the third generation of neural networks inspired by the biological neuron models [7] .  ... 
doi:10.1109/isvlsi.2016.91 dblp:conf/isvlsi/AhmedSWQ16 fatcat:hnozqognbzhazo3y2b3tarlvra

RRAM based neuromorphic algorithms [article]

Roshan Gopalakrishnan
2019 arXiv   pre-print
This report mainly talks about the work on deep neural network to spiking neural network conversion and its significance.  ...  This report basically gives an overview of the algorithms implemented on neuromorphic hardware with crossbar array of RRAM synapses.  ...  The integrate and fire (IF) neuron model was extensively used in SDNN until [30] demonstrated that a CNN can also be mapped onto a SDNN made up of leaky integrate and fire (LIF) neurons which are more  ... 
arXiv:1903.02519v1 fatcat:kjb5c4e5yfhkjb7cft4nwgfqfi
« Previous Showing results 1 — 15 out of 7,803 results