Among many artificial neural networks, the research on Spike Neural Network (SNN), which mimics the energy-efficient signal system in the brain, is drawing much attention. Memristor is a promising candidate as a synaptic component for hardware implementation of SNN, but several non-ideal device properties are making it challengeable. In this work, we conducted an SNN simulation by adding a device model with a non-linear weight update to test the impact on SNN performance. We found that SNN has a strong tolerance for the device non-linearity and the network can keep the accuracy high if a device meets one of the two conditions: 1. symmetric LTP and LTD curves and 2. positive non-linearity factors for both LTP and LTD. The reason was analyzed in terms of the balance between network parameters as well as the variability of weight. The results are considered to be a piece of useful prior information for the future implementation of emerging device-based neuromorphic hardware.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7996210 | PMC |
| http://dx.doi.org/10.3389/fncom.2021.646125 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Lateralization of Neural Speech Discrimination at Birth Is a Predictor for Later Language Development.
Dev Sci
March 2025
Department of Pediatrics and Adolescent Medicine, Comprehensive Center for Pediatrics, Medical University of Vienna, Vienna, Austria.
Newborns are able to neurally discriminate between speech and nonspeech right after birth. To date it remains unknown whether this early speech discrimination and the underlying neural language network is associated with later language development. Preterm-born children are an interesting cohort to investigate this relationship, as previous studies have shown that preterm-born neonates exhibit alterations of speech processing and have a greater risk of later language deficits.
View Article and Find Full Text PDFMolecular Mechanisms of Humic Acid in Inhibiting Silica Scaling during Membrane Distillation.
Environ Sci Technol
January 2025
Shandong Key Laboratory of Water Pollution Control and Resource Reuse, School of Environmental Science and Engineering, Shandong University, Qingdao 266237, P. R. China.
Membrane distillation (MD) efficiently desalinizes and treats high-salinity water as well as addresses the challenges in handling concentrated brines and wastewater. However, silica scaling impeded the effectiveness of MD for treating hypersaline water and wastewater. Herein, the effects of humic acid (HA) on silica scaling behavior during MD are systematically investigated.
View Article and Find Full Text PDFAssessment of the Accuracy of a Deep Learning Algorithm- and Video-based Motion Capture System in Estimating Snatch Kinematics.
Int J Exerc Sci
December 2024
Department of Sport and Health Sciences, Technical University of Munich, Munich, BY, GERMANY.
In weightlifting, quantitative kinematic analysis is essential for evaluating snatch performance. While marker-based (MB) approaches are commonly used, they are impractical for training or competitions. Markerless video-based (VB) systems utilizing deep learning-based pose estimation algorithms could address this issue.
View Article and Find Full Text PDFtdCoxSNN: Time-dependent Cox survival neural network for continuous-time dynamic prediction.
J R Stat Soc Ser C Appl Stat
January 2025
Department of Biostatistics and Health Data Science, School of Public Health, University of Pittsburgh, Pittsburgh, PA, USA.
The aim of dynamic prediction is to provide individualized risk predictions over time, which are updated as new data become available. In pursuit of constructing a dynamic prediction model for a progressive eye disorder, age-related macular degeneration (AMD), we propose a time-dependent Cox survival neural network (tdCoxSNN) to predict its progression using longitudinal fundus images. tdCoxSNN builds upon the time-dependent Cox model by utilizing a neural network to capture the nonlinear effect of time-dependent covariates on the survival outcome.
View Article and Find Full Text PDFSupplementary motor area in speech initiation: A large-scale intracranial EEG evaluation of stereotyped word articulation.
iScience
January 2025
Vivian L. Smith Department of Neurosurgery, McGovern Medical School at UT Health Houston, Houston, TX 77030, United States of America.
Speech production engages a distributed network of cortical and subcortical brain regions. The supplementary motor area (SMA) has long been thought to be a key hub in coordinating across these regions to initiate voluntary movements, including speech. We analyzed direct intracranial recordings from 115 patients with epilepsy as they articulated a single word in a subset of trials from a picture-naming task.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!