Memoization is a top-down dynamic-programming technique: begin with the original problem, recursively evaluate only the needed subproblems, and retain computed results for reuse. section 6.1.4 section 10.2.5
It evaluates the same recurrence, states, transitions, and base cases as tabulation; the methods differ in the order subproblems are encountered and stored. section 6.1.4
Multistage network optimization
A recurrence combines each choice’s immediate cost with the best outcome from the next state. Memoization can retain evaluated subproblems; with optimal substructure and exact evaluation, their solutions yield a global optimum. section 10.2.5
Its computational benefit depends on overlapping subproblems. Without optimal substructure, the global guarantee fails; without repeated subproblems, storing results offers little benefit. section 10.2.5








