| 1. | Convergence of new memory gradient method with armijo search
搜索下的记忆梯度法及其收敛性 |
| 2. | A class of conjugate gradient methods with armijo - type line searches
型线搜索下共轭梯度法簇的全局收敛性 |
| 3. | Convergence properties of the conjugate descent method with armijo - type line searches
型线搜索下共轭下降法的收敛性 |
| 4. | Global convergence results of a new three terms conjugate gradient method with generalized armijo step size rule
步长搜索的一类新的三项共轭梯度算法及其收敛特征 |
| 5. | In this chapter , we point out that the armijo - type line search will make the step k very small even close to zero in some cases , which leads to the pathological problem
在本章中，我们指出采用armijo型的线搜索，在某些情况下易使步长变的很小，甚至趋于零，这将在计算过程中导致病态现象。 |
| 6. | Based on this observation , we develop a newton - filter method in such a way that at each step , the method generates a point such that either the objective function value or its gradient descreases sufficiently
此时，常用的线性搜索(如armijo搜索和wolfe搜索等)可能失败。另一方面，不管目标函数的hessian阵是否正定， newton方向一定是目标函数的梯度的模函数的下降方向。 |
| 7. | Observing that the newton direction is always a descent direction of the square norm of the gradient , even if it is not a descent direction of the objective function , which will result some linear research - such as armijo research or wolfe research - fail
因此，该算法是颇受欢迎的算法之一。然而，若问题的目标函数的hessian阵不正定，则不能保证算法产生的方向是目标函数的下降方向。 |
