An Efficient Hybrid Optimization Algorithm Based on Lagged-Start and Parallel Operation

Fast convergence-rate, low computation complexity and good stability are important goals in the researching area of neural network learning algorithm. A kind of parallel computing lagged-start hybrid optimization algorithm is studied, it not only integrates the basic gradient method and the unconstrained optimization algorithm to realize the supplement of their advantages, but also makes full use of the high-performance computer´s parallel computing features to complete the algorithm switching from one to another on time, which improves the efficiency of algorithm learning and meets the neural network system´s online learning or real-time control. Combined a typical test function, a Microsoft Visual C# program is edit for the performance testing and validation of the proposed algorithm, the results is satisfied as expected.