相关实验视频

Updated: Jul 26, 2025

Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks

Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks

Published on: March 2, 2015

培训:最小化从卷积神经网络转换到尖端神经网络的准确性损失

Zhanglu Yan, Jun Zhou, Weng-Fai Wong

IEEE transactions on pattern analysis and machine intelligence

|June 14, 2023

概括

本研究介绍了CQ+培训,这是一种用于SNN的新算法,可以显著提高CIFAR数据集的准确性. 该方法可以有效地将卷积神经网络 (CNN) 转换为SNN,性能损失最小.

科学领域:

人工智能的人工智能
计算机视觉计算机视觉
神经形态计算是一种神经形态计算.

背景情况:

尖端神经网络 (SNN) 为边缘计算提供能源效率,因为它们的事件驱动性.
传统的卷积神经网络 (CNN) 达到更高的准确性,但更耗电.
弥合SNN和CNN之间的精度差距对于实际的SNN部署至关重要.

研究的目的:

开发一种与SNN兼容的训练算法,以提高SNN的准确性.
为了使各种CNN架构能够高效地转换为SNN.
为了减少SNNs中的延迟,而不会影响准确性.

主要方法:

拟为SNN提供CQ+培训 (扩展的"紧式"和"定量化"培训).
在实验中使用了经过修改的7层VGG模型 (VGG-*).
引入了参数化的输入编码和值培训,以缩短时间窗口.

主要成果:

在CIFAR-10上使用VGG-*与SNN转换 (0.09%的下降) 实现了最先进的95.06%准确性.
通过CQ+培训,时间窗口缩短到64,在CIFAR-10上保持94.09%的准确性.
在CIFAR-100上与VGG-*和500的时间窗口实现了77.27%的准确性.
已证明ResNet,MobileNet和Densenet的近零精度损失转换为具有小时间窗口 (<60) 的SNN.

结论:

CQ+培训有效地弥合了CNN和SNN之间的准确性差距.
拟议的方法使各种CNN能够高效准确地转换为SNN.
该PyTorch框架是公开可用的,促进了SNN的进一步研究和应用.

更多相关视频

A Simple Stimulatory Device for Evoking Point-like Tactile Stimuli: A Searchlight for LFP to Spike Transitions

A Simple Stimulatory Device for Evoking Point-like Tactile Stimuli: A Searchlight for LFP to Spike Transitions

Published on: March 25, 2014

Optical Recording of Suprathreshold Neural Activity with Single-cell and Single-spike Resolution

Optical Recording of Suprathreshold Neural Activity with Single-cell and Single-spike Resolution

Published on: September 5, 2012

相关概念视频

Improving Translational Accuracy

Improving Translational Accuracy

Base complementarity between the three base pairs of mRNA codon and the tRNA anticodon is not a failsafe mechanism. Inaccuracies can range from a single mismatch to no correct base pairing at all. The free energy difference between the correct and nearly correct base pairs can be as small as 3 kcal/ mol. With complementarity being the only proofreading step, the estimated error frequency would be one wrong amino acid in every 100 amino acids incorporated. However, error frequencies observed in...

Reducing Line Loss

Reducing Line Loss

In a three-phase circuit, line loss is an indicator of energy dissipated as heat due to the resistance of transmission lines. To address this, incorporating transformers into the system—a step-up transformer at the source and a step-down transformer at the load—is a strategic solution. Two three-phase transformers are introduced to improve this.
With a step-up transformer at the source, the voltage is increased, thereby reducing the current in the transmission lines since power loss...

Neural Circuits

Neural Circuits

Neural circuits and neuronal pools are two of the main structures found in the nervous system. Neural circuits are networks of neurons that work together to carry out a specific task or process. They consist of interconnected neurons and glial cells, which provide structural and metabolic support.
Neuronal pools are collections of nerve cells with similar functions and interact through chemical and electrical signals. These pools include both interneurons (the central neural circuit nodes that...