Biasing the quantum vacuum to control macroscopic probability distributions

Abstract

One of the most important insights of quantum field theory is that electromagnetic fields must fluctuate. Even in the vacuum state, the electric and magnetic fields have a nonzero variance, leading to ubiquitous effects such as spontaneous emission, the Lamb shift, the Casimir effect, and more. These "vacuum fluctuations" have also been harnessed as a source of perfect randomness, for example to generate perfectly random photonic bits. Despite these achievements, many potential applications of quantum randomness in fields such as probabilistic computing rely on controllable probability distributions, which have not yet been realized on photonic platforms. In this work, we show that the injection of vacuum-level "bias" fields into a multi-stable optical system enables a controllable source of "biased" quantum randomness. We demonstrate this concept in an optical parametric oscillator (OPO). Ordinarily, an OPO initiated from the ground state develops a signal field in one of two degenerate phase states (0 and $\pi$) with equal probability. By injecting bias pulses which contain less than one photon on average, we control the probabilities associated with the two output states, leading to the first controllable photonic probabilistic bit (p-bit). We shed light on the physics behind this process, showing quantitative agreement between theory and experiment. Finally, we demonstrate the potential of our approach for sensing sub-photon level fields by showing that our system is sensitive to the temporal shape of bias field pulses far below the single photon level. Our results suggest a new platform for the study of stochastic quantum dynamics in nonlinear driven-dissipative systems, and point toward possible applications in ultrafast photonic probabilistic computing, as well as the sensing of extremely weak fields.