Optimal-substructure
In
computer science, the
Optimal-substructure property refers to problems with optimal solutions that exhibit optimal solutions in its subproblems. This property is used to determine the usefulness of
dynamic programming and greedy algorithms in a problem.
An example of Optimal-substructure includes the fact that if a subproblem Sab has an activity Py, then it should contain optimal solutions to subproblems Say and Syb.
- TODO: Add Recursion, misc.
An optimal substructure problem is the
longest-common subsequence problem.