Black-box Importance Sampling
This addresses a bottleneck in machine learning and statistics for researchers and practitioners who rely on importance sampling for difficult problems.
The paper tackles the limitation of importance sampling being restricted to simple proposals by introducing a black-box method that works with any unknown proposal, enabling the use of richer proposals and improving estimation accuracy.
Importance sampling is widely used in machine learning and statistics, but its power is limited by the restriction of using simple proposals for which the importance weights can be tractably calculated. We address this problem by studying black-box importance sampling methods that calculate importance weights for samples generated from any unknown proposal or black-box mechanism. Our method allows us to use better and richer proposals to solve difficult problems, and (somewhat counter-intuitively) also has the additional benefit of improving the estimation accuracy beyond typical importance sampling. Both theoretical and empirical analyses are provided.