Polynomial-Time Reducible
#Computers
To show a problem is NP complete (not easy), you may show that a known NP complete problem reduces into the given problem
$\displaystyle Y\leq _p X\land Y\text{ is NP complete}\rightarrow Y\text{ is NP complete}$
- Reads that if $\displaystyle Y$ is polynomial reducible/transformable to $\displaystyle X$ and $\displaystyle Y$ is NP complete, then Y must be NP complete
- Proof by contradiction:
- Assume $\displaystyle Y$ isn't NP complete with the above givens
- Then we could solve $\displaystyle X$ in polynomial time by solving $\displaystyle Y$ and then reducing it to $\displaystyle X$, which contradicts the fact that $\displaystyle X$ is NP complete
| $\displaystyle Y$ | $\displaystyle X$ | $\displaystyle X\leq_{p}Y$ too? | Description |
|---|---|---|---|
| Maximum Clique Problem | Maximum Independent Set Problem | Yes | Invert all edges/non-edges in $\displaystyle O(n^{2})$ time |
| Maximum Independent Set Problem | Vertex Cover Problem | Yes | Take set of nodes that aren't in VC as MIS |
| Vertex Cover Problem | Efficient Recruiting Problem | Yes | From $\displaystyle G(V,E)$, let edges represent sports and nodes counselors |