Cook's theorem

Cook's theorem

Cook's theorem, proved by Stephen Cook in 1971, established that the Boolean satisfiability problem (SAT) is NP-complete, linking problems studied by Alan Turing, John von Neumann, Alonzo Church, Kurt Gödel, and groups at Bell Labs. The result spurred research at institutions such as MIT, Stanford University, Princeton University, and University of Toronto and influenced awards like the Gödel Prize and projects such as the P vs NP problem initiatives. It connects to work by contemporaries including Richard Karp, Leonid Levin, Donald Knuth, Stephen Cook (computer scientist), and influenced conferences like STOC and FOCS.

Background

The context for Cook's theorem arises from early 20th-century developments by Alan Turing and Alonzo Church on decidability, and mid-20th-century investigations by John von Neumann and Kurt Gödel on computability. Studies at University of California, Berkeley and Princeton University on algorithmic complexity and formal models led to formal classes such as P (complexity class) and NP (complexity class), and to the articulation of the P versus NP problem. Research groups at Bell Labs and universities including Harvard University, Columbia University, and University of Toronto explored reductions and completeness, while subsequent expositions by Richard Karp and Leonid Levin disseminated the notion of NP-completeness across conferences like STOC and FOCS.

Statement of the theorem

Cook's theorem asserts that the Boolean satisfiability problem (SAT) is NP-complete: SAT belongs to NP (complexity class) and every problem in NP (complexity class) can be polynomial-time reduced to SAT. The theorem formalizes the notion of NP-hardness and NP-completeness introduced in the milieu of work by Stephen Cook, building on the formal machine model of Alan Turing and the logical frameworks of Alonzo Church and Kurt Gödel. It implies that if any NP (complexity class)-complete problem such as SAT is in P (complexity class), then landmark conjectures like the P versus NP problem and results affecting cryptography studied at RSA Laboratories and institutions like Bell Labs would be overturned.

Proof outline

Cook's original proof constructs, for an arbitrary nondeterministic Turing machine M and input x, a Boolean formula that is satisfiable exactly when M accepts x within a polynomial bound. The construction encodes the machine's configuration tableau using symbols and states from models developed by Alan Turing and formalized by Alonzo Church, employs polynomial-time simulation techniques akin to approaches in works by John Hopcroft and Jeffrey Ullman, and uses reductions familiar from later expositions by Richard Karp. Key steps: represent tape contents, head position, and state at each time step; enforce transition constraints via clauses; and bound time by a polynomial function linked to results in computational complexity from MIT and Stanford University. Variants of the encoding appear in textbooks by Michael Sipser and Christos Papadimitriou and are taught in courses across Harvard University and Princeton University.

Consequences and significance

Cook's theorem precipitated the formal study of NP-completeness, motivating Karp's 1972 list of 21 NP-complete problems and influencing domains such as cryptography at RSA Laboratories, algorithm design at Bell Labs, and theoretical research at IBM Research and Microsoft Research. It created a common framework linking problems in graph theory (research in Graph Theory (disambiguation) and studies involving Erdős–Rényi model), combinatorial optimization studied at INFORMS-affiliated groups, and decision problems in logic traced to Kurt Gödel and Alonzo Church. The theorem underpins complexity-theoretic assumptions used in protocols by National Institute of Standards and Technology and shapes research programs at conferences like STOC and FOCS and journals including those of ACM and SIAM.

Related concepts and developments

Developments related to Cook's theorem include Richard Karp's reductions, Leonid Levin's independent characterization of NP-completeness, and the ongoing P versus NP problem which remains central at institutions including Clay Mathematics Institute and in prizes such as the Millennium Prize Problems. Subsequent refinements produced classes like co-NP (complexity class), PSPACE (complexity class), and completeness notions such as PSPACE-complete and parameterized complexity explored by researchers at Carnegie Mellon University and University of Edinburgh. Practical branches influenced by the theorem include SAT solving research at Microsoft Research and industrial work at Google and Amazon on constraint solving, as well as computational complexity curricula at Massachusetts Institute of Technology and Stanford University.

Category:Theorems in theoretical computer science