Isoholonomic inequality and tight implementations of holonomic quantum gates

Abstract

In holonomic quantum computation, quantum logic gates are realized by cyclic parallel transport of the computational space. The resulting quantum gate corresponds to the holonomy associated with the closed path traced by the computational space. The isoholonomic inequality for gates establishes a fundamental lower bound on the path length of such cyclic transports, which depends only on the spectrum of the holonomy, that is, the eigenvalues of the implemented quantum gate. The isoholonomic inequality also gives rise to an estimate of the minimum time required to execute a holonomic quantum gate, underscoring the central role of the inequality in quantum computation. In this paper, we show that when the codimension of the computational space is sufficiently large, any quantum gate can be implemented using a parallel transporting Hamiltonian in a way that saturates the isoholonomic inequality and the corresponding time estimate. We call such implementations tight. The treatment presented here is constructive and lays the foundation for the development of efficient and optimal implementation strategies in holonomic quantum computation.