Data-driven Sequential Monte Carlo in Probabilistic Programming
This addresses efficiency bottlenecks in probabilistic programming for researchers and practitioners, representing an incremental improvement over existing methods.
The authors tackled the problem of suboptimal proposal distributions in probabilistic programming systems by training a neural network to approximate optimal proposals using posterior estimates from previous inference runs. Their results show data-driven proposals significantly improve inference performance, requiring considerably fewer particles for good posterior estimation.
Most of Markov Chain Monte Carlo (MCMC) and sequential Monte Carlo (SMC) algorithms in existing probabilistic programming systems suboptimally use only model priors as proposal distributions. In this work, we describe an approach for training a discriminative model, namely a neural network, in order to approximate the optimal proposal by using posterior estimates from previous runs of inference. We show an example that incorporates a data-driven proposal for use in a non-parametric model in the Anglican probabilistic programming system. Our results show that data-driven proposals can significantly improve inference performance so that considerably fewer particles are necessary to perform a good posterior estimation.