Spike Counts based Low Complexity SNN Architecture with Binary Synapse.

08:00 EDT 4th October 2019 | BioPortfolio

Summary of "Spike Counts based Low Complexity SNN Architecture with Binary Synapse."

In this paper, we present an energy and area efficient spike neural network (SNN) processor based on novel spike counts based methods. For the low cost SNN design, we propose hardware-friendly complexity reduction techniques for both of learning and inferencing modes of operations. First, for the unsupervised learning process, we propose a spike counts based learning method. The novel learning approach utilizes pre- and post-synaptic spike counts to reduce the bit-width of synaptic weights as well as the number of weight updates. For the energy efficient inferencing operations, we propose an accumulation based computing scheme, where the number of input spikes for each input axon is accumulated without instant membrane updates until the pre-defined number of spikes are reached. In addition, the computation skip schemes identify meaningless computations and skip them to improve energy efficiency. Based on the proposed low complexity design techniques, we design and implement the SNN processor using 65nm CMOS process. According to the implementation results, the SNN processor achieves 87.4% of recognition accuracy in MNIST dataset using only 1-bit 230k synaptic weights with 400 excitatory neurons. The energy consumptions are 0.26pJ/SOP and 0.31μJ/inference in inferencing mode, and 1.42pJ/SOP and 2.63μJ/learning in learning mode of operations.


Journal Details

This article was published in the following journal.

Name: IEEE transactions on biomedical circuits and systems
ISSN: 1940-9990


DeepDyve research library

PubMed Articles [15382 Associated PubMed Articles listed on BioPortfolio]

MorphIC: A 65-nm 738k-Synapse/Formula: see text Quad-Core Binary-Weight Digital Neuromorphic Processor with Stochastic Spike-Driven Online Learning.

Recent trends in the field of neural network accelerators investigate weight quantization as a means to increase the resource- and power-efficiency of hardware devices. As full on-chip weight storage ...

A Probabilistic Synapse With Strained MTJs for Spiking Neural Networks.

Spiking neural networks (SNNs) are of interest for applications for which conventional computing suffers from the nearly insurmountable memory-processor bottleneck. This paper presents a stochastic SN...

Sparse model-based estimation of functional dependence in high-dimensional field and spike multiscale networks.

Behavior is encoded across multiple scales of brain activity, from binary neuronal spikes to continuous fields including local field potentials (LFP). Multiscale models need to describe both the encod...

A complete statistical model for calibration of RNA-seq counts using external spike-ins and maximum likelihood theory.

A fundamental assumption, common to the vast majority of high-throughput transcriptome analyses, is that the expression of most genes is unchanged among samples and that total cellular RNA remains con...

Reading-out task variables as a low-dimensional reconstruction of neural spike trains in single trials.

We propose a new model of the read-out of spike trains that exploits the multivariate structure of responses of neural ensembles. Assuming the point of view of a read-out neuron that receives synaptic...

Clinical Trials [3541 Associated Clinical Trials listed on BioPortfolio]

Effect of Auditory Stimulation on Spike Waves in Sleep

Background: Close relationship exists between sleep slow wave (SSW) and the generation of spike wave in NREM-sleep. SSW are cortically generated oscillations alternating between excitatory...

Altered Connections in the Spinal Cord to Reduce Hand Impairment After Stroke

The goals of this study are to leverage principles of spike timing-dependent plasticity (STDP) via non-invasive stimulation techniques to strengthen corticospinal transmission. Few studies...

Sleep Related Memory Consolidation in Children With Age Related Focal Epilepsy.

Age related focal epilepsies in children encompasses, according to the ILAE criteria, benign epilepsy with centro temporal spikes (BECTS), atypical benign partial epilepsy (ABPE) and epile...

Efficacy of Flywheel Inertial Resistance Training in the Architecture and Function of the Hamstring

Investigate the efficacy of the inercial Flywheel training protocol in modifying the architecture and function of the hamstrings in patients with a history of structural muscle injury vs c...

Efficacy Study of Acetazolamide Versus Diazepam in Continuous Spike and Wave/Landau-Kleffner Syndrome

The purpose of this study is to compare the effectiveness of the medications acetazolamide and diazepam in the treatment of continuous spike wave in sleep (CSWS) and Landau-Kleffner syndro...

Medical and Biotech [MESH] Definitions

The art and science of designing buildings and structures. More generally, it is the design of the total built environment, including town planning, urban design, and landscape architecture.

Works consisting of drawings of architecture and architectural projects, whether the project was executed or not. (Art & Architecture Thesaurus, 1990, v.2)

Works on the art and science of designing buildings and structures. More generally, works on the design of the total built environment, including town planning, urban design, and landscape architecture.

Architecture, exterior and interior design, and construction of facilities other than hospitals, e.g., dental schools, medical schools, ambulatory care clinics, and specified units of health care facilities. The concept also includes architecture, design, and construction of specialized contained, controlled, or closed research environments including those of space labs and stations.

Inorganic binary compounds of iodine or the I- ion.

Quick Search

DeepDyve research library

Searches Linking to this Article