Comments for Computational Complexity, Spring 2010

Comment on Homework #3 by anuprao

anuprao — Wed, 12 May 2010 06:44:55 +0000

Hi Gilbert, you are right, the size of the set should depend on epsilon. In general you can find a set of size poly(s, 1/epsilon). Also, I meant for P to be a multiset (or else the size may depend on n). I will send an email about this problem soon.

Comment on Homework #3 by Gilbert Bernstein

Gilbert Bernstein — Wed, 12 May 2010 05:21:19 +0000

On 3, does epsilon depend on anything? If not, then doesn’t this force the two probabilities to be equal? Doesn’t this further permit the construction of counterexamples for small circuits? (say quadratic size)

Comment on Homework #3 by anuprao

anuprao — Tue, 04 May 2010 03:48:46 +0000

Yes Rolfe, you are right. It is now fixed. Thanks!

Comment on Homework #3 by Rolfe

Rolfe — Tue, 04 May 2010 02:57:53 +0000

In problem 2 is it supposed to be ?

Comment on Homework #1 by Thach Nguyen

Thach Nguyen — Wed, 07 Apr 2010 20:14:08 +0000

It’s be great if you can turn them in via emails. That way, I can be sure not to lose any of them .

Also, email me if you need to talk about any of the question.

Thanks,
Thach