I am getting my head around convexity theory for machine learning and am looking into finding the right hyperparameter for a parametrized regression model/deep learning model. I know that you can calculate the hyperparameters through theory, it can tell us how to set hyperparameters for strongly convex optimization, which is defined like this,
Strong Convexity - a function $f$ is $\alpha$-strongly convex if $$ \nabla^2 f(x) \succeq \alpha I $$ for all $x$, where $I$ is the identity matrix, if we know the strong convexity constant µ and the Lipschitz constant L.
How to get the Lipschitz constant and strong convexity constant from strict convex problems to calculate hyperparameters in this case? Is it even useful to calcualte the hyperparameters through theory or is it normally done experimentally by using some convex loss function like least-square?