A life spent making mistakes is not only more honorable, but more useful than a life spent doing nothing.
George Bernard Shaw
As we have seen throughout this book, techniques for solving constraint problems fall into two main categories: search (also called conditioning) and inference. Search and inference can also be combined, trading off time and space, and improving overall performance (see Chapter 10). Nevertheless, even when combined, their worst-case time complexity is often prohibitive, requiring the use of approximations.
Greedy search is one of the most appealing approximation methods. One reason for this method’s popularity is that it requires very little memory. A popular greedy approach ...