Branch and Bound
In plain terms
An optimization search that prunes branches that can't beat the best found.
Going deeper
Branch-and-bound explores a solution tree but discards subtrees whose best possible outcome is worse than the current best. It's used for hard optimization problems.
Under the hood
Bounds estimate the best achievable in a branch, cutting the search dramatically versus brute force.
Examples
Traveling salesman, integer programming, knapsack, scheduling.
Think of it like...
Like house-hunting but skipping whole neighborhoods once you know they can't beat your current best find.