Computational complexity and 3-manifolds and zombies
From MaRDI portal
Publication:6288926
DOI10.2140/GT.2018.22.3623arXiv1707.03811WikidataQ129181549 ScholiaQ129181549MaRDI QIDQ6288926FDOQ6288926
Authors: Greg Kuperberg, Eric Samperton
Publication date: 12 July 2017
Abstract: We show the problem of counting homomorphisms from the fundamental group of a homology -sphere to a finite, non-abelian simple group is #P-complete, in the case that is fixed and is the computational input. Similarly, deciding if there is a non-trivial homomorphism is NP-complete. In both reductions, we can guarantee that every non-trivial homomorphism is a surjection. As a corollary, for any fixed integer , it is NP-complete to decide whether admits a connected -sheeted covering. Our construction is inspired by universality results in topological quantum computation. Given a classical reversible circuit , we construct so that evaluations of with certain initialization and finalization conditions correspond to homomorphisms . An intermediate state of likewise corresponds to a homomorphism , where is a pointed Heegaard surface of of genus . We analyze the action on these homomorphisms by the pointed mapping class group and its Torelli subgroup . By results of Dunfield-Thurston, the action of is as large as possible when is sufficiently large; we can pass to the Torelli group using the congruence subgroup property of . Our results can be interpreted as a sharp classical universality property of an associated combinatorial -dimensional TQFT.
Computational difficulty of problems (lower bounds, completeness, difficulty of approximation, etc.) (68Q17) Finite-type and quantum invariants, topological quantum field theories (TQFT) (57K16) Invariants of 3-manifolds (including skein modules, character varieties) (57K31)
This page was built for publication: Computational complexity and 3-manifolds and zombies
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6288926)