基于区间优化的神经网络全局优化方法

A Neural Network Global Optimization Algorithm Based on Interval Optimization

Hopfield神经网络被广泛应用于优化问题的求解中,而传统的Hopfield网络通常基于梯度下降法,此方法容易陷入局部极小而得到次最优解或收敛到问题的不可行解。另外,当用于训练网络样本的输入/输出数据无法精确给出,而只能以一定的范围的形式给出时,传统的神经网络学习方法就无能为力了。论文提出了一种基于区间优化的神经网络学习算法,可以很好地解决上面所提到的传统神经网络学习算法的缺点。

Hopfield neural networks are often used for optimization problems.But they are usually based on gradientbased procedures.Such methods are frequently find sub-optimal solutions being trapped in local minima or find an unfeasible solution.It will be difficult for the traditional neural network when the input/output data sets used to train a neural network may not be hundred percent precise but within certain range.A learning algorithm based on interval optimization is presented in this paper.The above disadvantages of the traditional learning algorithm can be settled by using this method。