The optimizers are lifted from convex optimization, but the point above was that they are applied to highly non-convex problems. They work for finding local min

The optimizers are lifted from convex optimization, but the point above was that they are applied to highly non-convex problems. They work for finding local minima, but a lot of the deeper literature does not translate (e.g. the conjecture being discussed in this post).