Why Le Cam equation is not popular but very useful ? [R]

I recently got interested in Le Cam equation which informally is useful to find the relevant metric in the hypothesis class/ collection of densities in order to find (sub)linear rate of convergence for the minimax estimation of error.
I will leave some references and I’m interested to know more about a geometrical interpretation to understand the intuition behind it. I’m interested in work where this approach was used, particularly in bandits, I would also be grateful if you can recommend …

collection convergence equation error hypothesis linear machinelearning minimax popular rate will