Generalization ability is a major problem encountered when using neural networks to find the structures in noisy data sets. Controlling the network complexity is a common method to solve this problem. In this paper, h...Generalization ability is a major problem encountered when using neural networks to find the structures in noisy data sets. Controlling the network complexity is a common method to solve this problem. In this paper, however, a novel additive penalty term which represents the features extracted by hidden units is introduced to eliminate the overtraining of multilayer feedfoward networks. Computer simulations demonstrate that by using this unsupervised fashion penalty term, the generalization ability is greatly improved.展开更多
The authors investigate the stability of a steady ideal plane flow in an arbitrary domain in terms of the L^2 norm of the vorticity. Linear stability implies nonlinear instability provided the growth rate of the line...The authors investigate the stability of a steady ideal plane flow in an arbitrary domain in terms of the L^2 norm of the vorticity. Linear stability implies nonlinear instability provided the growth rate of the linearized system exceeds the Liapunov exponent of the flow. In contrast,a maximizer of the entropy subject to constant energy and mass is stable. This implies the stability of certain solutions of the mean field equation.展开更多
文摘Generalization ability is a major problem encountered when using neural networks to find the structures in noisy data sets. Controlling the network complexity is a common method to solve this problem. In this paper, however, a novel additive penalty term which represents the features extracted by hidden units is introduced to eliminate the overtraining of multilayer feedfoward networks. Computer simulations demonstrate that by using this unsupervised fashion penalty term, the generalization ability is greatly improved.
文摘The authors investigate the stability of a steady ideal plane flow in an arbitrary domain in terms of the L^2 norm of the vorticity. Linear stability implies nonlinear instability provided the growth rate of the linearized system exceeds the Liapunov exponent of the flow. In contrast,a maximizer of the entropy subject to constant energy and mass is stable. This implies the stability of certain solutions of the mean field equation.