摘要 :
This paper presents a quantum-based algorithm for evolving artificial neural networks (ANNs). The aim is to design an ANN with few connections and high classification performance by simultaneously optimizing the network structure ...
展开
This paper presents a quantum-based algorithm for evolving artificial neural networks (ANNs). The aim is to design an ANN with few connections and high classification performance by simultaneously optimizing the network structure and the connection weights. Unlike most previous studies, the proposed algorithm uses quantum bit representation to codify the network. As a result, the connectivity bits do not indicate the actual links but the probability of the existence of the connections, thus alleviating mapping problems and reducing the risk of throwing away a potential candidate. In addition, in the proposed model, each weight space is decomposed into subspaces in terms of quantum bits. Thus, the algorithm performs a region by region exploration, and evolves gradually to find promising subspaces for further exploitation. This is helpful to provide a set of appropriate weights when evolving the network structure and to alleviate the noisy fitness evaluation problem. The proposed model is tested on four benchmark problems, namely breast cancer and iris, heart, and diabetes problems. The experimental results show that the proposed algorithm can produce compact ANN structures with good generalization ability compared to other algorithms.
收起
摘要 :
This article studies spacecraft rendezvous with the target spacecraft in an arbitrary elliptical orbit. Based on the linearised Tschauner??Hempel equations, state-dependent Riccati differential equation (RDE)-based approaches are ...
展开
This article studies spacecraft rendezvous with the target spacecraft in an arbitrary elliptical orbit. Based on the linearised Tschauner??Hempel equations, state-dependent Riccati differential equation (RDE)-based approaches are proposed to solve the problem. An observer-based output feedback controller is established. To implement the proposed controller, only some forward-type RDEs need to be integrated online, indicating that the controllers can be implemented online as long as some positive-definite initial conditions are given. Moreover, it is shown analytically that the closed-loop system under the proposed feedback controller is exponentially asymptotically stable. Numerical examples show the effectiveness of the proposed approach.
收起
