Transfer learning has attracted more and more attention, and many scholars proposed some useful strategies. Boosting is the main strategy for transfer learning. In boosting, resampling is preferred over reweighting, a...
详细信息
ISBN:
(纸本)9781509061839
Transfer learning has attracted more and more attention, and many scholars proposed some useful strategies. Boosting is the main strategy for transfer learning. In boosting, resampling is preferred over reweighting, and it can be applied to any base learner. In this paper, we propose a weighted-resampling method for transfer learning, called TrResampling. Firstly, resampling is applied to the data with heaven weight in the source domain, and the resampled data is used with the target data as the training data to build a classifier. Then the TrAdaBoost algorithm is used to adjust the weights of source data and target data. We discuss Decision Tree, Naive Bayes, and SVM as the base learner in TrResampling, and choose the suitable for TrResampling. In order to illustrate the performance of the proposed algorithm, we compare TrResampling with the state-of-the-art algorithm TrAdaBoost and the base learner Decision Tree, experimental results on UCI data sets indicate that TrResampling is superior to TrAdaBoost and Decision Tree on many data sets.
In this paper, the external consensus problem of networked multi-agent systems with fixed undirected topology, random network delay and nonlinear dynamics is considered. The unknown nonlinear dynamics existing in the ...
详细信息
ISBN:
(纸本)9781509015740;9781509015733
In this paper, the external consensus problem of networked multi-agent systems with fixed undirected topology, random network delay and nonlinear dynamics is considered. The unknown nonlinear dynamics existing in the systems can be described as RBF-ARX model. Besides, the random network delay can be compensated by introducing the prediction strategy. The RBF-ARX model is used as prediction model to design the output predictor, which is to generate the prediction sequence of control input and agent's output. Then, the designed selector chooses the proper value from the available sequences corresponding to the value of the network delay. Based on the mentioned above, a distributed external consensus control algorithm is proposed to make the considered systems with external reference input achieve consensus. Finally, an example is given to illustrate the validity of the proposed algorithm.
A new disturbance rejection method to achieves good control performance for a nonlinear time-delay system is presented in this paper. This method effectively reject any disturbances. First, the configuration of the co...
详细信息
ISBN:
(纸本)9781509015740;9781509015733
A new disturbance rejection method to achieves good control performance for a nonlinear time-delay system is presented in this paper. This method effectively reject any disturbances. First, the configuration of the controlsystem is constructed based on the equivalent-input-disturbance method. Then, a sufficient condition for global uniform ultimate boundedness and feedback controller are developed using linear matrix inequality. Finally, the effectiveness of the method is proved by a numerical simulation.
The content based image retrieval (CBIR) is aimed to find the most similar images from a collection of images or a database to the query image according to the visual or semantic similarity. Current image retrieval al...
详细信息
作者：
Gong KunDeng FangMa TaoGong Kun is with School of Automation
Beijing Institute of Technology and Key Laboratory of Advanced Control of Iron and Steel Process (Ministry of Education) Beijing China Deng Fang is with School of Automation
Beijing Institute of Technology and Key Laboratory of Advanced Control of Iron and Steel Process (Ministry of Education) Beijing China Ma Tao is with School of Automation
Beijing Institute of Technology and Key Laboratory of Complex System Intelligent Control and Decision Ministry of Education Beijing China
In order to improve the precision of the azimuth measured by mobile robot's electronic compass, this paper proposes a new calibration method based on Fourier Neural Network trained by Modified Particle Swarm Optim...
详细信息
In order to improve the precision of the azimuth measured by mobile robot's electronic compass, this paper proposes a new calibration method based on Fourier Neural Network trained by Modified Particle Swarm Optimization (MPSO-FNN). This method makes use of Fourier Neural Network (FNN) to establish the error compensation model of electronic compass's azimuth, and introduces Modified Particle Swarm Optimization (MPSO) algorithm to optimize the weights of neural network. Thus the comparatively accurate error model of azimuth is obtained to compensate the output of electronic compass. This method not only has strong nonlinear approximation capability, but also overcomes the neural networks' shortcomings which are too slow convergence speed, oscillation, and easy to fall into local optimum and sensitive to the initial values. Experimental results demonstrate that after calibrated by this method, the range of azimuth error reduces to -0.35°~0.70° from -3.4°~25.2°, and the average value of absolute error is only 0.30°.
This paper is devoted to robust adaptive sliding mode control for a class of nonlinear systems in the Takagi-Sugeno forms with mismatched parametric uncertainties. Sufficient conditions for the existence of linear sli...
详细信息
暂无评论