Could somebody describe the PCP Theorem in layman’s terms? [edit:Im a computer science student beginning theory.]
Could somebody describe the PCP Theorem in layman’s terms? [edit:Im a computer science student
Share
Sign Up to our social questions and Answers Engine to ask questions, answer people’s questions, and connect with other people.
Login to our social questions & Answers Engine to ask questions answer people’s questions & connect with other people.
Lost your password? Please enter your email address. You will receive a link and will create a new password via email.
Please briefly explain why you feel this question should be reported.
Please briefly explain why you feel this answer should be reported.
Please briefly explain why you feel this user should be reported.
Could somebody describe the PCP Theorem in layman’s terms? [edit:Im a computer science student beginning theory.]
Targeted towards the layman computer scientist:
The PCP theorem says that you can make proofs that are so easy to check that you only need to look at a constant number of (randomly selected) bits to (usually) tell a bad proof from a good one.
Targeted towards the layman theorist:
For a language to be in NP means that you can prove a string is in the language easily. For example, you can prove a boolean circuit is satisfiable by providing its satisfying input. (but it’s hard to prove a circuit is not satisfiable!)
In this context the person/program/turing machine verifying the proof has to run in polynomial time (in the size of the string supposedly in the language). It must be complete, meaning that it always accepts a valid proof a given string is in the language. It must be sound, meaning it rejects any ‘proof’ for a string that is not in the language.
For NP the verifier is deterministic. What if the verifier can use randomness? We then have to allow for some error, so we allow the verifier to not be totally sound— it can improperly accept some proofs. Really, all we care about is that the verifier detects bad proofs at least, say, 1/4 of the time. If you want better accuracy you can always run it a few more times! 🙂
The PCP theorem, in its strongest form, says that any language in NP has a proof system that can be verified by only looking at a constant number of randomly selected bits. The constant is… 3. (That’s the same 3 from 3-SAT) These proofs use error correcting codes, such that any error in the proof will conflict with a large portion of the input. Using a stronger code lets you get arbitrarily close to 50% soundness with just 3 bits.