Robust quantum computation by simulation

Simulation of quantum systems that provide intrinsically fault-tolerant quantum computation is shown to preserve fault tolerance. Errors committed in the course of simulation are eliminated by the natural error-correcting features of the systems simulated. Two examples are explored, toric codes and non-abelian anyons. The latter is shown to provide universal robust quantum computation via simulation. Quantum computers are devices that process information in a way that preserves quantum coherence [1-8]. Because of the ubiquity of decohering processes and the difficulty of performing logic operations at the scale of atoms, photons, etc., quantum computations are more sensitive to noise and errors than classical computations [5-10]. A method for performing quantum computation is called fault-tolerant if it is intrinsically resistant to noise and errors committed in the course of computation; such a method is termed robust if it allows the accurate performance of arbitrarily long quantum computations in the presence of a finite error rate [8-10]. Recently, two methods for performing robust quantum computation have been proposed. The first relies on the theory of quantum error correction codes, and uses quantum logic to correct both errors introduced by noise, and errors introduced in the course of performing quantum logic itself [8-10]. The second method proposes a class of physical systems whose dynamics are automatically fault tolerant: in such systems, quantum information is stored on topological excitations that are immune to local errors [11-15]. This paper proposes a third method for performing robust quantum computation that is intermediate between these two methods: quantum logic is used to