Learning Optimal Flows for Non-Equilibrium Importance Sampling

Many applications in computational sciences and statistical inference require the computation of expectations with respect to complex high-dimensional distributions with unknown normalization constants, as well as the estimation of these constants. Here we develop a method to perform these calculations based on generating samples from a simple base distribution, transporting them along the flow generated by a velocity field, and performing averages along these flowlines. This non-equilibrium importance sampling (NEIS) strategy is straightforward to implement, and can be used for calculations with arbitrary target distributions. On the theory side we discuss how to tailor the velocity field to the target and establish general conditions under which the proposed estimator is a perfect estimator, with zero-variance. We also draw connections between NEIS and approaches based on mapping a base distribution onto a target via a transport map. On the computational side we show how to use deep learning to represent the velocity field by a neural network and train it towards the zero variance optimum. These results are illustrated numerically on high dimensional examples, where we show that training the velocity field can decrease the variance of the NEIS estimator by up to 6 order of magnitude compared to a vanilla estimator. We also show that NEIS performs better on these examples than Neal's annealed importance sampling (AIS).