Overview
A polynomial-time reduction transforms instances of one problem, called problem A, into instances of another problem, called problem B, using an algorithm that runs in polynomial time. For a decision-preserving reduction, the transformation must preserve the yes-or-no answer: an instance is a yes-instance of A if and only if its transformed instance is a yes-instance of B.
Key concepts
name: Source Problem; description: The original problem whose instances are transformed.; name: Target Problem; description: The problem into which the original instances are transformed.; name: Polynomial-Time Transformation; description: The conversion must be computable in time bounded by a polynomial in the input size.; name: Answer Preservation; description: The transformation must preserve the relevant answer or solution relationship.; name: Reduction Direction; description: The direction of a reduction determines which hardness conclusions can be established.
Example
scenario: Using a known NP-complete problem to establish the hardness of another problem.; source: 3-SAT.; target: A graph-based decision problem.; approach: Construct graph vertices and edges from the variables and clauses of a 3-SAT instance so that the graph has a specified structure exactly when the formula is satisfiable.; purpose: A correct polynomial-time construction proves that the target problem is at least as hard as 3-SAT.; importantNote: The exact graph construction depends on the target problem. A reduction must prove the equivalence rather than merely show that the problems appear similar.
Applications
Proving NP-hardness and NP-completeness.; Comparing the difficulty of decision problems.; Transferring algorithmic results between related problems.; Studying approximation and optimization problems.; Organizing theoretical computer science research.
Limitations
The suitability of polynomial-time reductions depends on the exact problem requirements and the assumptions made about the input.; An implementation of polynomial-time reductions can fail if its boundary conditions, data representation, or required invariants are handled incorrectly.; For polynomial-time reductions, evaluate representative and unusually difficult inputs; a single small demonstration is not enough to establish reliability.
Common mistakes
Reversing the direction of a reduction.; Showing only one direction of the answer-preservation condition.; Ignoring the time required to construct the transformed instance.; Assuming that a reduction automatically provides an efficient solution.; Confusing a polynomial-time reduction with an algorithm that solves the original problem directly.