Nelder–Mead has often not worked well for me in moderate to high dimensions. I'd recommend trying Powell's method if you want to quickly converge to a local opt

Nelder–Mead has often not worked well for me in moderate to high dimensions. I'd recommend trying Powell's method if you want to quickly converge to a local optimum. If you're using scipy's wrappers, it's easy to swap between the two: https://docs.scipy.org/doc/scipy/reference/optimize.html#local-multivariate-optimization https://docs.scipy.org/doc/scipy/reference/optimize.html#loc... For nastier optimization problems there are lots of other options, including evolutionary algorithms and Bayesian optimization: https://facebookresearch.github.io/nevergrad/ https://facebookresearch.github.io/nevergrad/ https://github.com/facebook/Ax https://github.com/facebook/Ax