Title:
Fusion-assisted decoding of non-Abelian topological order
Abstract:
Maximum likelihood decoders perform decoding on noisy quantum states by measuring a set of syndromes, classically computing an optimal recovery based on these syndromes, and then performing that recovery. For the toric code state under incoherent Pauli noise, the maximum likelihood decoder using syndromes corresponding to the local anyon excitations reaches the intrinsic mixed-state information theoretic recoverability threshold, which maps onto a point along the Nishimori line of the random-bond Ising model. For non-abelian topological orders, maximum likelihood decoders using local anyon syndromes do not in general reach information theoretic thresholds. We show that using enlarged syndrome sets that include measurements that resolve non-abelian anyon fusion outcomes leads to improved maximum likelihood decoders. A hierarchy of increasingly refined syndrome sets gives rise to a hierarchy of effective mixed states from the perspective of the decoder. We study this hierarchy concretely by constructing explicit decoders for the $D_4$ quantum double model under charge noise, and we obtain numerical results for thresholds via mappings to classical statistical-mechanics models.
Bio: