I'm not a student, I'm watching these lectures as kind of self learner. I had a very hard time understanding the part of #SAT-IP proof at the end, when Ryan didn't have any more time to further show the details. Several sources did a very bad job of explaining, surprisingly wikipedia had a really good explanation of the core mechanism: en.wikipedia.org/wiki/IP_%28complexity%29##SAT_is_a_member_of_IP Scroll to "Phase i", it really shows how it works, in every round the prover sends a polynomial, the verifier sends a number a_i, the polynomial of the current round has to work with the sum (over the b_i=0,1) and the result of that sum is to be the same as the result of the polynomial of the round before, when you insert all the a_i into it.
Thanks so much for the very clear lesson and explanation!
I'm not a student, I'm watching these lectures as kind of self learner. I had a very hard time understanding the part of #SAT-IP proof at the end, when Ryan didn't have any more time to further show the details. Several sources did a very bad job of explaining, surprisingly wikipedia had a really good explanation of the core mechanism:
en.wikipedia.org/wiki/IP_%28complexity%29##SAT_is_a_member_of_IP
Scroll to "Phase i", it really shows how it works, in every round the prover sends a polynomial, the verifier sends a number a_i, the polynomial of the current round has to work with the sum (over the b_i=0,1) and the result of that sum is to be the same as the result of the polynomial of the round before, when you insert all the a_i into it.