郑重声明:原文参见标题,如有侵权,请联系作者,将会撤销发布!
https://arxiv.org/abs/2305.16044
Summary
Keywords
Introduction
Results
Noisy spiking neural network and noise-driven learning
NSNN leads to high-performance spiking neural models
NSNN leads to improved robustness against challenging perturbations
NSNN demonstrates a promising tool for neural coding research
Discussion
EXPERIMENTAL PROCEDURES
Data and code availability
Method details
Notations
LIF neuron
Noisy LIF neuron
Noisy Spiking Neural Network
Noise-driven Learning
Theoretical analyses on internal noise and stability
Experimental details
Details of recognition experiments
Effect of internal noise level on network performance
Details of recognition experiments with perturbations
Details of recognition task coding analyses
Details of neural activity fitting experiments
Tables, table titles, and table legends
- Computation Exploiting Learning Networks Resourcecomputation exploiting learning networks learning networks neural hello superglue learning matching networks pre-train learning networks neural networks learning neural comp sequence learning networks neural quot demonstrated networks learning convolutional importance learning networks exploiting computation