A co-evolutional immune algorithm for the optimization of a function with real parameters is de-scribed.It uses a cooperative co-evolution of two populations,one is a population of antibodies and theother is a populat...A co-evolutional immune algorithm for the optimization of a function with real parameters is de-scribed.It uses a cooperative co-evolution of two populations,one is a population of antibodies and theother is a population of successful mutation vectors.These two population evolve together to improve thediversity of the antibodies.The algorithm described is then tested on a suite of optimization problems.The results show that on most of test functions,this algorithm can converge to the global optimum atquicker rate in a given range,the performance of optimization is improved effetely.展开更多
An incremental time-delay neural network based on synapse growth, which is suitable for dynamic control and learning of autonomous robots, is proposed to improve the learning and retrieving performance of dynamical re...An incremental time-delay neural network based on synapse growth, which is suitable for dynamic control and learning of autonomous robots, is proposed to improve the learning and retrieving performance of dynamical recurrent associative memory architecture. The model allows steady and continuous establishment of associative memory for spatio-temporal regularities and time series in discrete sequence of inputs. The inserted hidden units can be taken as the long-term memories that expand the capacity of network and sometimes may fade away under certain condition. Preliminary experiment has shown that this incremental network may be a promising approach to endow autonomous robots with the ability of adapting to new data without destroying the learned patterns. The system also benefits from its potential chaos character for emergence.展开更多
基金Supported by the National Fundamental Research Project(A1420060159)
文摘A co-evolutional immune algorithm for the optimization of a function with real parameters is de-scribed.It uses a cooperative co-evolution of two populations,one is a population of antibodies and theother is a population of successful mutation vectors.These two population evolve together to improve thediversity of the antibodies.The algorithm described is then tested on a suite of optimization problems.The results show that on most of test functions,this algorithm can converge to the global optimum atquicker rate in a given range,the performance of optimization is improved effetely.
文摘An incremental time-delay neural network based on synapse growth, which is suitable for dynamic control and learning of autonomous robots, is proposed to improve the learning and retrieving performance of dynamical recurrent associative memory architecture. The model allows steady and continuous establishment of associative memory for spatio-temporal regularities and time series in discrete sequence of inputs. The inserted hidden units can be taken as the long-term memories that expand the capacity of network and sometimes may fade away under certain condition. Preliminary experiment has shown that this incremental network may be a promising approach to endow autonomous robots with the ability of adapting to new data without destroying the learned patterns. The system also benefits from its potential chaos character for emergence.
