Open Confusio opened 6 months ago
In my problem, the evaluation of the target function and the associated gradient has noise and computational errors. In the near future, is it possible to include the Stochastic Quasi-Newton Method in Optim?
I doubt it. Do you have a specific thing in mind?
In my problem, the evaluation of the target function and the associated gradient has noise and computational errors. In the near future, is it possible to include the Stochastic Quasi-Newton Method in Optim?