Annealed Importance Sampling with q-Paths

Abstract

Annealed importance sampling (AIS) is the gold standard for estimating partition functions or marginal likelihoods, corresponding to importance sampling over a path of distributions between a tractable base and an unnormalized target. While AIS yields an unbiased estimator for any path, existing literature has been primarily limited to the geometric mixture or moment-averaged paths associated with the exponential family and KL divergence. We explore AIS using $q$-paths, which include the geometric path as a special case and are related to the homogeneous power mean, deformed exponential family, and $\alpha$-divergence.

arXiv commentNeurIPS Workshop on Deep Learning through Information Geometry (Best Paper Award)

References in corpus (5)

• Equilibrium free energy differences from nonequilibrium measurements: a master equation approach
• An elementary introduction to information geometry
• Sandwiching the marginal likelihood using bidirectional Monte Carlo
• Measuring the reliability of MCMC inference with bidirectional Monte Carlo
• Advantages of $q$-logarithm representation over $q$-exponential representation from the sense of scale and shift on nonlinear systems

Cited by in corpus (2)

• $\alpha$-Geodesical Skew Divergence
• Parallel Tempering on Optimized Paths