Normally in machine learning we use derivatives in-order to optimize a cost function . But I saw monte carlo optimization method which uses integrals in-order to optimize specially in reinforcement learning problems and some sampling problems ?
What is the difference of these two types ?
Best Answer
Derivatives and Monte-Carlo techniques are not used in the same way for optimization.
If a cost function is easy to compute we can use derivatives to perform a gradient descent method of optimizing parameters.
If a cost function is expensive to compute then it might not be feasible to calculate the derivatives. In these cases we must be careful with how many times we compute the cost function, we need an efficient method of optimizing the parameters for the cost function.
A Monte-Carlo approach is used for estimation, not for optimization. However, estimation is helpful for producing an efficient optimization algorithm. If the cost function is expensive to compute then we can use Monte-Carlo methods to estimate it with much less computing time.