A Variant of the VC-dimension with Applications to Depth-3 Circuits

11/18/2021
βˆ™
by   Peter Frankl, et al.
βˆ™
0
βˆ™

We introduce the following variant of the VC-dimension. Given S βŠ†{0, 1}^n and a positive integer d, we define π•Œ_d(S) to be the size of the largest subset I βŠ† [n] such that the projection of S on every subset of I of size d is the d-dimensional cube. We show that determining the largest cardinality of a set with a given π•Œ_d dimension is equivalent to a TurΓ‘n-type problem related to the total number of cliques in a d-uniform hypergraph. This allows us to beat the Sauer–Shelah lemma for this notion of dimension. We use this to obtain several results on Ξ£_3^k-circuits, i.e., depth-3 circuits with top gate OR and bottom fan-in at most k: * Tight relationship between the number of satisfying assignments of a 2-CNF and the dimension of the largest projection accepted by it, thus improving Paturi, Saks, and Zane (Comput. Complex. '00). * Improved Ξ£_3^3-circuit lower bounds for affine dispersers for sublinear dimension. Moreover, we pose a purely hypergraph-theoretic conjecture under which we get further improvement. * We make progress towards settling the Ξ£_3^2 complexity of the inner product function and all degree-2 polynomials over 𝔽_2 in general. The question of determining the Ξ£_3^3 complexity of IP was recently posed by Golovnev, Kulikov, and Williams (ITCS'21).

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset
Success!
Error Icon An error occurred

Sign in with Google

×

Use your Google Account to sign in to DeepAI

×

Consider DeepAI Pro