Comments on: Lower bounds by negative adversary method http://cstheory.blogoverflow.com/2011/08/lower-bounds-by-negative-adversary-method/ The Theoretical Computer Science Stack Exchange Blog Fri, 18 Mar 2016 21:21:18 +0000 hourly 1 https://wordpress.org/?v=4.5.6 By: Artem Kaznatcheev http://cstheory.blogoverflow.com/2011/08/lower-bounds-by-negative-adversary-method/#comment-111 Wed, 28 Sep 2011 17:26:03 +0000 http://cstheory.blogoverflow.com/?p=96#comment-111 Why does the expression for certificate complexity hold only for total functions? I was a bit informal in my descriptions, by not specifying what x and y range over, but it is implicitly assumed (as you see in the duals) that they range over the domain of the function. Thus the expression for certificate complexity should hold for partial functions as well.

I definitely agree that the ‘general adversary bound’ is a better name than the ‘negative adversary method’. I have used the latter mostly because I have learnt it as such, and because many people are already familiar with the (positive) adversary method. However, I will make a conscious effort to refer to it as the ‘general adversary bound’ from now on.

]]> By: Anonymous http://cstheory.blogoverflow.com/2011/08/lower-bounds-by-negative-adversary-method/#comment-107 Mon, 26 Sep 2011 17:25:54 +0000 http://cstheory.blogoverflow.com/?p=96#comment-107 Your expression for certificate complexity holds only for total functions. The “negative adversary method” is more properly known as the “general adversary bound.”

]]>