> the RL training Is RL necessary in this context or would other, simpler methods work as well to within statistical error? RL can be a heck of a random number

> the RL training Is RL necessary in this context or would other, simpler methods work as well to within statistical error? RL can be a heck of a random number generator if misapplied.