As a new neural network model,extreme learning machine(ELM)has a good learning rate and generalization ability.However,ELM with a single hidden layer structure often fails to achieve good results when faced with large...As a new neural network model,extreme learning machine(ELM)has a good learning rate and generalization ability.However,ELM with a single hidden layer structure often fails to achieve good results when faced with large-scale multi-featured problems.To resolve this problem,we propose a multi-layer framework for the ELM learning algorithm to improve the model’s generalization ability.Moreover,noises or abnormal points often exist in practical applications,and they result in the inability to obtain clean training data.The generalization ability of the original ELM decreases under such circumstances.To address this issue,we add model bias and variance to the loss function so that the model gains the ability to minimize model bias and model variance,thus reducing the influence of noise signals.A new robust multi-layer algorithm called ML-RELM is proposed to enhance outlier robustness in complex datasets.Simulation results show that the method has high generalization ability and strong robustness to noise.展开更多
An in-depth study of the complicated trajectory characteristics of skipping stones is carried out in the present work.A three-dimensional numerical simulation validated by acquiring good agreement with experimental re...An in-depth study of the complicated trajectory characteristics of skipping stones is carried out in the present work.A three-dimensional numerical simulation validated by acquiring good agreement with experimental results is established.It is devoted to illustrating five different types of motion responses after the stone impacts the water surface in the o-xy plane,including“dive”,“hydroplaning trout”,“hydroplaning skip”,“stable skip”,and“skipping trout”.Then,the lateral deviations are investigated quantitatively based on dimensionless parameter sin(α+β)cosαin the o-yz plane.Steady interval and linear interval are divided for lateral deviation Z1/D and Z2/D based on the values of(α+β)andα,respectively.The results reveal that(1)Z1/D increases almost linearly with the increasing sin(α+β)cosαat different slopes in different(α+β)intervals;(2)Z2/D increases almost linearly with the increasing sin(α+β)cosαat different slopes in differentαintervals;(3)in linear interval,numerical lateral deviations are much larger than the fitting values at pointsβ/α≥4.5,and much smaller than the fitting values at points 2.3≤β/α<4.3.Finally,a theoretical approach is proposed to predict the maximum immersion depth of trailing edge point P.展开更多
基金Project(21878081)supported by the National Natural Science Foundation of ChinaProject(222201917006)supported by the Fundamental Research Funds for the Central Universities,China。
文摘As a new neural network model,extreme learning machine(ELM)has a good learning rate and generalization ability.However,ELM with a single hidden layer structure often fails to achieve good results when faced with large-scale multi-featured problems.To resolve this problem,we propose a multi-layer framework for the ELM learning algorithm to improve the model’s generalization ability.Moreover,noises or abnormal points often exist in practical applications,and they result in the inability to obtain clean training data.The generalization ability of the original ELM decreases under such circumstances.To address this issue,we add model bias and variance to the loss function so that the model gains the ability to minimize model bias and model variance,thus reducing the influence of noise signals.A new robust multi-layer algorithm called ML-RELM is proposed to enhance outlier robustness in complex datasets.Simulation results show that the method has high generalization ability and strong robustness to noise.
基金supported by the National Natural Science Foundation of China(Grant No.11972138).
文摘An in-depth study of the complicated trajectory characteristics of skipping stones is carried out in the present work.A three-dimensional numerical simulation validated by acquiring good agreement with experimental results is established.It is devoted to illustrating five different types of motion responses after the stone impacts the water surface in the o-xy plane,including“dive”,“hydroplaning trout”,“hydroplaning skip”,“stable skip”,and“skipping trout”.Then,the lateral deviations are investigated quantitatively based on dimensionless parameter sin(α+β)cosαin the o-yz plane.Steady interval and linear interval are divided for lateral deviation Z1/D and Z2/D based on the values of(α+β)andα,respectively.The results reveal that(1)Z1/D increases almost linearly with the increasing sin(α+β)cosαat different slopes in different(α+β)intervals;(2)Z2/D increases almost linearly with the increasing sin(α+β)cosαat different slopes in differentαintervals;(3)in linear interval,numerical lateral deviations are much larger than the fitting values at pointsβ/α≥4.5,and much smaller than the fitting values at points 2.3≤β/α<4.3.Finally,a theoretical approach is proposed to predict the maximum immersion depth of trailing edge point P.
