LM algorithm is between Newton and gradient descent between a nonlinear optimization method, the parameters of the problem is not too sensitive, and can effectively deal with the issue redundant parameters, the objective function into a local minimum of the opportunity to greatly reduce the small

LM algorithm is midway between Newton's law and the gradient between the decline of a non-linear optimization method, for being too parameterized problems, are not sensitive to effectively address issues redundant parameters in which the target function into a local minimum value significantly reduc

The LM algorithm is situated between between the Newton law and the gradient drop law one non-linear optimization method, is insensitive regarding the too parametrization question, can process the redundant parametric problem effectively, causes the objective function to fall into the local minimum

LM algorithm is between Newton and a nonlinear optimization method of gradient descent method, for being too insensitive parametric problems, can effectively handle redundant parameter problem that causes the target function to a local minimum of opportunity to significantly reduce

LM algorithm is a non-linear optimization method between Newton law and gradient descent method, as to overly parameter insensitivity of question, can deal with the redundant parameter problem effectively, the chance of making the goal function fall into some minimum is greatly reduced