期刊文献+

快速自顶向下优化神经网络结构的方法 被引量:4

Fast Approach for Optimal Brain Surgeon
下载PDF
导出
摘要 引荐了一种优化神经网络结构的改进方法。近年来,各种神经网络结构优化方法相继被提出,但素负盛名的还是神经中枢手术优化算法(OBS)。提出一种与修剪技术相关的快速方法,这种方法起始于适当复杂的初始网络,然后剔除冗余的神经元。研究说明,这种快速方法与以往的常规神经中枢手术优化算法相比,具有更好的归纳性能、更简单的网络结构和更快的学习速度。 An improved pruning algorithm is presented. Although various pruning algorithms have been developed, possibly the most popular of these is optimal brain surgeon (OBS). The way used by the improved algorithm is to train an initial network that is reasonable large and then eliminate the unneeded parts. Experiment results of comparative studies with formal OBS show that the improved pruning algorithm provides improvement of generalization capability, and reduction of both neural network complexity and training time.
机构地区 西北工业大学
出处 《系统仿真学报》 EI CAS CSCD 北大核心 2005年第1期162-165,共4页 Journal of System Simulation
关键词 神经中枢手术优化算法 修剪 结构优化 自顶向下 反向传播算法 Optimal Brain Surgeon Pruning Architecture Optimization Top-down Backpropagation
  • 相关文献

参考文献15

  • 1哈根(Hagan M T)等 戴葵等译.神经网络设计[M].北京:机械工业出版社,2002..
  • 2Haykin S. Neural Networks, A Comprehensive Foundation [M]. New York: Macmillan, 1994.
  • 3Fletcher R. Practical Methods of Optimization [M]. New York: Wiley,1990.
  • 4Rumelhart D E , Hmton G E , Williams R J . Learning Internal Representations by Error Propagation [A]. Rumelhart D E ,McClelland J L . Parallel Distributed Processing: Explanations in the Microstmcture of Cognition [C]. Cambridge, MA: MIT Press,1986,1:318-362.
  • 5Kung S, Fallside F, Sorenson J A, Kamm C. Neural Networks for Signal Processing [C]. Proceeding of the 1992 IEEE Workshop,1992, Ⅱ:255-266.
  • 6Guyon I, Wang P S P. Special Issue on Neural Networks and Pattern Recognition[J]. Pattern Recognition Artificial Intelligence, 1993, 7(4):849-872.
  • 7Friedman J H . An Overview of Predictive Learning and Function Approximation [A]. Friedman J H , Wechsler H . From Statistics to Neural Networks: Theory and Pattern Recognition Applications,Proceeding of the ASI Conference, Subseries F[C]. New York:Springer-Vedag, 1994.
  • 8Moody J . Prediction Risk and Architecture Selection for Neural Networks [A]. Cherkassky V , Friedman J ,Wechsler H . From Statistics to Neural Networks: Theory and Pattern Recognition Applications, NATO ASI Series F [C]. New York: Springer-Verlag,1994, 136: 147-165.
  • 9Fahlman S E , Lcbicrc C. The Cascade-Correlation Learning Architecture [A]. Touretzky D S . Advances in Neural Information Processing Systems [C]. San Mateo, CA: Morgan Kaufmann, 1990, 2:524-532.
  • 10Reed R. Pruning Algorithms -- A Survey [J].IEEE Transaction on Neural Networks, 1993, 4(5): 740-747.

共引文献1

同被引文献43

引证文献4

二级引证文献10

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部