Traditional optimization methods use much more information of the target problem, so their convergence speed is much better, and the ability of finding local optimal is better.

传统的优化方法充分利用了目标问题的信息，局部寻优能力较强，收敛速度较快，但又会陷入局部最优的陷阱。

youdao

The traditional training methods of Gaussian Mixture Model(GMM) are sensitive to the initial model parameters, which often leads to a local optimal parameter in practice.

为了解决传统高斯混合模型（GMM）对初值敏感，在实际训练中极易得到局部最优参数的问题，提出了一种采用微粒群算法优化GMM参数的新方法。

youdao