Frequency-splitting estimators for quark-line disconnected diagrams

In this talk I will present recent work on improved estimators for quark-line disconnected diagrams in lattice QCD.
The evaluation of these diagrams is required for many phenomenologically interesting observables, but suffers from large statistical errors due to vacuum contributions.
Multi-level integration schemes ameliorate such signal-to-noise ratio problems and have the potential to allow exponentially-improved precision in the computation of these observables.
To best exploit the multi-level for disconnected diagrams, the dominant noise due to auxiliary stochastic fields should be reduced below the gauge noise.
We have performed a theoretical analysis of the variances which has lead to new improved estimators, applicable with or without multi-level integration.
I will discuss the prospects of using such estimators in the context of the hadronic vacuum polarization contribution to the muon anomalous magnetic moment.