From the course: AI Algorithms for Game Design with Python

Unlock this course with a free trial

Join today to access over 24,800 courses taught by industry experts.

Writing good evaluation functions

Writing good evaluation functions - Python Tutorial

From the course: AI Algorithms for Game Design with Python

Start my 1-month free trial Buy for my team

Writing good evaluation functions

“

- [Instructor] So here's what depth limited search does. We select a constant level at which we wish to stop the search, and this will leave us with a very valuable smaller tree. Notice that we will not analyze most of the original tree, and that's fine. The running time for the algorithm on this tree depends on the level at which we are willing to stop. So some trial and error may help us decide this level. However, this improvement in time comes with a cost, and this cost is accuracy. At the bottom of this smaller tree, we need a good evaluation function to estimate how well we'll do if we take the paths underneath these bottom nodes. If we use a poor evaluation function, our agent will behave poorly. Remember that Minimax produces the best solution possible, so that's the best we could aim for, and we must be ready to have a less than perfect agent. In order to minimize these imperfections, we need to come up with a good evaluation function. Here we have three important details to…

Contents