List Agreement Expansion from Coboundary Expansion
share
One of the key components in PCP constructions are agreement tests. In agreement test the tester is given access to subsets of fixed size of some set, each equipped with an assignment. The tester is then tasked with testing whether these local assignments agree with some global assignment over the entire set. One natural generalization of this concept is the case where, instead of a single assignment to each local view, the tester is given access to l different assignments for every subset. The tester is then tasked with testing whether there exist l global functions that agree with all of the assignments of all of the local views. In this work we present sufficient condition for a set system to exhibit this generalized definition of list agreement expansion. This is, to our knowledge, the first work to consider this natural generalization of agreement testing. Despite initially appearing very similar to agreement expansion, list agreement expansion seem to require a different set of techniques. This is due to the fact that the natural extension of agreement testing does not suffice when testing for list agreement, as list agreement crucially relies on a global structure. It follows that if a local assignments satisfy list agreement they must not only agree locally but also exhibit some additional structure. In order to test for the existence of this additional structure we use a connection between covering spaces of a high dimensional complex and its coboundaries. We use this connection as a form of “decoupling”. Moreover, we show that any set system that exhibits list agreement expansion also supports direct sum testing. This is the first scheme for direct sum testing that works regardless of the parity of the sizes of the local sets. Prior to our work the schemes for direct sum testing were based on the parity of the sizes of the local tests.
READ FULL TEXT