In order to improve the generalization capability of stochastic configuration networks(SCNs),a smooth L1 regularization method for SCNs is proposed.Aiming at the defect of local non-differentiability of the L1 regularization operator,smoothing is carried out in the neighborhood of non-smooth points of the curve.The convex error function of the SCN is constructed on this basis,and an algorithm for incremental calculation of the weights of the SCN is proposed.Furthermore,the global optimization algorithm is proposed based on the alternating direction multiplier method,and the convergence of the algorithm is analyzed theoretically.Compared with the sparsity of L1 regularization and the uniform reduction of parameters by L2 regularization,the proposed method retains all features of the data according to the degree of importance,the parameters are not only kept in a small range,but also have hierarchical distribution,so that the network has better generalization ability.Finally,the feasibility and effectiveness of the proposed method are verified by some numerical simulations.
关键词
光滑正则化/随机配置网络/泛化能力/交替方向乘子法/收敛性分析/数据特征
Key words
smoothing regularization/stochastic configuration networks/generalization capability/alternating direction multiplier method/convergence analysis/data feature
维普
万方数据