在1983 年,Sejnowski, T和. Hinton, G. E 设计玻尔兹曼机,首次提出了隐层单元概念。
在1986年,Geoffrey E. Hinton, Ronald J. Williams, David E. Rumelhart 提出反向传播的算法[5]。在反向传播算法之后,提出径向基神经网络,简称RBF (RadialBasis Function Neural Network)[6]。这种神经网络仅仅由三层组成,包括输入层,隐藏层,输出层。非线性变化应用到输入层和隐藏层之间,线性变化应用到隐藏层和输出层之间。RBF算法采用核函数的思想,将低维不可分的向量映射到高维度。使得向量在低维度不可分时,在高维度可分,提高模型对分类任务的性能。
在2006 年,G. E. Hinton 等人提出深度信念网络, 简称DBN (Deep Belief Network)[8]。
1998 年,Yann Le Cun使用卷积神经网络对于字符进行识别,提出LeNet5[10]。
2012 年,AlexNet在ImageNet 图像分类竞赛中取得优异的性能[11]。
2017年,Vaswani A 使用注意力机制应用在自然语言处
参考资料
[1] MCCULLOCH W S, PITTS W H. A Logical Calculus of the Ideas Immanent in Nervous Activity[G] //BODEN M A. Oxford readings in philosophy : The Philosophy of Artificial Intelligence. :Oxford University Press, 1990 : 22 – 39. [2] HEBB D O. The organization of behavior: A neuropsychological theory[M]. : Psychology Press,2005. [3] ROSENBLATT F. The perceptron: a probabilistic model for information storage and organization in the brain.[J]. Psychological review, 1958, 65(6) : 386. [4] BLOCK H D. A Review of ”Perceptrons: An Introduction to Computational Geometry” by Marvin Minsky and Seymour Papert[J/OL]. Inf. Control., 1970, 17(5) : 501 – 522. https://doi.org/10.1016/ S0019-9958(70)90409-2. [5] RUMELHART D E, HINTON G E, WILLIAMS R J. Learning representations by back-propagating errors[J]. nature, 1986, 323(6088) : 533 – 536. [6] BROOMHEAD D S, LOWE D. Radial basis functions, multi-variable functional interpolation and adaptive networks[R]. : Royal Signals and Radar Establishment Malvern (United Kingdom), 1988. [7] HEARST M A, DUMAIS S T, OSUNA E, et al. Support vector machines[J]. IEEE Intelligent Systems and their applications, 1998, 13(4) : 18 – 28. [8] HINTON G E. Deep belief networks[J]. Scholarpedia, 2009, 4(5) : 5947. [9] EDDY S R. What is a hidden Markov model?[J]. Nature biotechnology, 2004, https://blog.csdn.net/jinking01/article/details/103344186