Codes for "Differential estimates for fast multilevel nonconvex optimisation"

These are the Julia codes for the numerical demonstrations of the manuscript“Differential estimates for fast multilevel nonconvex optimisation” (arXiv:2412.01481).

Prerequisites

These codes were written for Julia 1.12. You may get Julia from julialang.org.

Using

Navigate your unix shell to the directory containing this README.md and then run:

$ julia --project=.

The first time doing this, to ensure all the dependencies are installed, run

]instantiate

Afterwards in the Julia shell, type:

using TrackingCodes

This may take a while as Julia precompiles the code. Then, to generate all the experiments in the manuscript, run:

batchrun_demos()

You will find the results in the directory res/. To only run the EIT or minimal surface experiments, you can use:

TrackingCodes.EIT.batchrun_demos()
TrackingCodes.MinSurf.batchrun_demos()