10. Dynamic programming (DP)

Dynamic programming (DP) is a method for solving complex problems by breaking them down into smaller subproblems, solving each of those subproblems just once, and storing their solutions. The key idea behind dynamic programming is to avoid redundant calculations by using the results of previous computations to solve subsequent ones. This technique is particularly useful when the same or similar subproblems are encountered repeatedly during the solution process.

Dynamic programming can be applied to problems that have two important properties:

1. Optimal Substructure: An optimal solution to the problem contains optimal solutions to its subproblems. In other words, the optimal solution to the whole problem can be constructed from the optimal solutions of its parts.
2. Overlapping Subproblems: There must be overlapping subproblems in the sense that the solution to a given subproblem will be needed multiple times during the calculation of the final result.

The general approach to solving a problem with dynamic programming involves three main steps:

1. Define the State: Identify the “state” of the problem, which represents the information necessary to determine the value of the solution for any given subproblem.
2. Recursively Define the Value Function: Express the value of an optimal solution for a given state as a function of the values of optimal solutions for substates. This relationship is often defined through a recurrence relation or recursive formula.
3. Compute the Optimal Solution: Use the computed values to construct an optimal solution to the original problem. This step typically involves building up the solution from the bottom up, based on the stored subproblem solutions.

Dynamic programming has applications across various fields such as computer science, operations research, economics, and engineering. It is commonly used in optimization problems, including but不限于背包问题、最短路径问题、序列对齐问题、股票 price prediction, and many others.