Flow Annealed Importance Sampling Bootstrap
This addresses a bottleneck in computational physics and machine learning for sampling from unnormalized densities, offering a more efficient method for molecular systems like alanine dipeptide.
The paper tackles the problem of training normalizing flows to approximate complex multimodal distributions like Boltzmann distributions without relying on expensive pre-generated samples or suffering from high variance, achieving results that outperform maximum likelihood training with 100 times fewer target evaluations and producing unbiased histograms nearly identical to ground truth.
Normalizing flows are tractable density models that can approximate complicated target distributions, e.g. Boltzmann distributions of physical systems. However, current methods for training flows either suffer from mode-seeking behavior, use samples from the target generated beforehand by expensive MCMC methods, or use stochastic losses that have high variance. To avoid these problems, we augment flows with annealed importance sampling (AIS) and minimize the mass-covering $α$-divergence with $α=2$, which minimizes importance weight variance. Our method, Flow AIS Bootstrap (FAB), uses AIS to generate samples in regions where the flow is a poor approximation of the target, facilitating the discovery of new modes. We apply FAB to multimodal targets and show that we can approximate them very accurately where previous methods fail. To the best of our knowledge, we are the first to learn the Boltzmann distribution of the alanine dipeptide molecule using only the unnormalized target density, without access to samples generated via Molecular Dynamics (MD) simulations: FAB produces better results than training via maximum likelihood on MD samples while using 100 times fewer target evaluations. After reweighting the samples, we obtain unbiased histograms of dihedral angles that are almost identical to the ground truth.