Modeling dynamic evolution of continual state spatial field based on CA and AR method;

Aiming at the reinforcement learning controller design under continuous state space,a new Q-learning method based on a support vector machine(SVM) was proposed.

We used this method to solve a typical inventory control problem with continuous state and decision space.

Denoising method of anisotropic diffusion based oncontinuous state wavelet threshold

In this paper,a class of strong limit theorem about the non-homogeneous Markov chain taking values inthe continuous state are obtained in virtue of the notion of likelihood ratio and martingale convergence theorem.

The reliability of the system which consists of continuous state components is discussed in this paper.

For reinforcement learning control in continuous spaces,a Q-learning method based on a self-organizing fuzzy RBF(radial basis function) network is proposed.

This paper is to study genetic algorithm in the unified framework of stochastic processes in continuous space.

To overcome the multiple redundancy and low efficient solving of genetic algorithms as a result of no feedback ability, as well as low speed of ant colony algorithms owing to absence of original pheromone, a modified ant colony algorithm has been improved, which is capable of searching in continuous space, to integrate with genetic algorithm to complement each other’s advantages.

In this paper, the convergent conditions in sequence or parallel update mode and the sufficient condition with only one global stable state for Hopfield network model with discrete time and continuous states when its neurons?activation function is non-decreasing (not being strictly monotone increasing) are discussed.