论文信息 - A multi-network architecture for high generalization in pattern recognition with backpropagation neural network modules

A multi-network architecture for high generalization in pattern recognition with backpropagation neural network modules

Backpropagation networks are the most popular multi-layer networks, used for either function approximation or pattern classification. They are trained and tested using two disjoint sets of patterns drawn randomly from the pattern space. In many cases, the overtraining phenomenon occurs i.e. the network learns to produce the proper output for the patterns to which it has been trained but it produces meaningless outputs for unforeseen patterns. In this paper, the overtraining phenomenon is analyzed in depth, and an alternative architecture with increased generalization ability is proposed.

Spyros G. Tzafestas | Y. Anthopoulos

[1] David Haussler,et al. Decision Theoretic Generalizations of the PAC Model for Neural Net and Other Learning Applications , 1992, Inf. Comput..

[2] Eric B. Baum,et al. When Are k-Nearest Neighbor and Back Propagation Accurate for Feasible Sized Sets of Examples? , 1990, EURASIP Workshop.

[3] Geoffrey E. Hinton,et al. Learning internal representations by error propagation , 1986 .

[4] Anders Krogh,et al. Introduction to the theory of neural computation , 1994, The advanced book program.

[5] Spyros G. Tzafestas,et al. On the overtraining phenomenon of backpropagation neural networks , 1996 .

[6] Vladimir Vapnik,et al. Chervonenkis: On the uniform convergence of relative frequencies of events to their probabilities , 1971 .