Decisions, Decisions: Noise and its Effects on Integral Monte Carlo Algorithms

In the present paper we examine the effects of noise on Monte Carlo algorithms, a problem raised previously by Kennedy and Kuti (Phys. Rev. Lett. {\bf 54}, 2473 (1985)). We show that the effects of introducing unbiased noise into the acceptance/rejection phase of the conventional Metropolis approach are surprisingly modest, and, to a significant degree, largely controllable. We present model condensed phase numerical applications to support these conclusions.