Chapter 9
Optimisation and Search
In almost all of the algorithms that we’ve looked at in the previous chapters there has been some element of optimisation, generally by defining some sort of error function, and attempting to minimise it. We’ve talked about gradient descent, which is the optimisation method that forms the basis of many machine learning algorithms. In this chapter, we will look at formalising the gradient descent algorithm and understanding how it works, and then we will look at what we can do when there are no gradients in the problem, and so gradient descent doesn’t work.
Whatever method we have used to solve the optimisation problem, the basic methodology has been the same: to compute the derivative of the error function to ...
Get Machine Learning, 2nd Edition now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.