you may find David Goodwin's new paper regarding Accelerated Newton-Raphson GRAPE methods for optimal control, which shows very impressive acceleration in the computation of the Hessian, in a large number of time steps. So do you have a plan to include the new features in this paper to Spinach?
I'm excited to give this method a try ASAP since LBFGS taking its time to converge in my simulation. But of course, it all depends on your schedule. Many thanks!
I had a good look - the code David had sent would require a complete rewrite to fit into the current GRAPE function in Spinach. I might eventually be able to do that, but the time scale is months.
Try contacting David, he was a co-author of our GRAPE module and understands how it works. He probably has more time on his hands and would be able to do that faster.
