相关实验视频

Updated: Jan 18, 2026

Deep Neural Networks for Image-Based Dietary Assessment

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

深度神经网络的基于能源的本地学习的有效方法和框架

Haibo Chen¹, Bangcheng Yang¹, Fucun He¹

¹China Nanhu Academy of Electronics and Information Technology, Jiaxing, China.

Frontiers in artificial intelligence

|September 11, 2025

概括

这项研究引入了一种基于能源的新型模型,可以克服预测编码 (PC) 的局限性,用于训练人工神经网络. 新方法实现了与反向传播相比较的高精度,同时提高了培训效率.

科学领域:

神经科学是一个神经科学.
人工智能的人工智能
机器学习机器学习

背景情况:

人工神经网络 (ANN) 经常使用生物学上不合理的反向传播进行训练.
基于能源的模型 (EBM) 提供了一个受大脑启发的替代方案,通过最小化能量功能来学习.
预测编码 (PC) 是一种EBM,它使用预测错误,但在深度网络中与梯度问题作斗争.

研究的目的:

在深度预测编码网络中解决梯度爆炸和消失问题.
提高大脑启发的学习模型的训练效率和稳定性.
为ANN开发一种具有竞争力的反向传播替代方案.

主要方法:

引入双向能量以稳定预测错误并减轻梯度爆炸.
实现了跳过连接来解决梯度消失问题.
开发了一种层级适应性学习率 (LALR),以提高培训效率.
创建了一个基于JAX的框架,以提供高效的EBM培训.

主要成果:

取得的高精度: 99.22% (MNIST),93.78% (CIFAR-10),83.96% (CIFAR-100) 和73.35% (Tiny ImageNet). 图像的精度是可以使用的.
性能可与以反向传播训练的相同网络相比较.
与使用新的Jax框架的PyTorch相比,培训时间减少了50%.

结论:

拟议的基于能源的双向模型有效地克服了PC在深度网络中的局限性.
这种由大脑启发的方法为反向传播提供了一个可行的和高效的替代方案.
开发的框架加速了EBM培训,展示了实际优势.

关键词:

人工神经网络的人工神经网络生物学上可信的学习规则.基于能源的模型当地学习学习当地学习预测编码的预测编码.

更多相关视频

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

相关概念视频

Introduction to Learning

Introduction to Learning

Learning is the process of acquiring knowledge or skills through practice or experience, leading to long-lasting behavioral changes. This acquisition occurs through interaction with the environment and requires practice or experience. For instance, mastering a skill such as surfing requires considerable practice and experience, highlighting the essential role of repeated interactions with the environment in learning.
In contrast to learned behaviors, unlearned behaviors such as crying, sexual...

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...