Divide, Conquer, and Combine: a New Inference Strategy for Probabilistic Programs with Stochastic Support
This addresses a critical bottleneck in probabilistic programming for researchers and practitioners, enabling more efficient automated inference for complex models, though it is an incremental improvement within the field.
The paper tackled the problem of inefficient inference in universal probabilistic programming systems for models with varying support, introducing a new inference framework called Divide, Conquer, and Combine that achieved substantial performance improvements over existing approaches.
Universal probabilistic programming systems (PPSs) provide a powerful framework for specifying rich probabilistic models. They further attempt to automate the process of drawing inferences from these models, but doing this successfully is severely hampered by the wide range of non--standard models they can express. As a result, although one can specify complex models in a universal PPS, the provided inference engines often fall far short of what is required. In particular, we show that they produce surprisingly unsatisfactory performance for models where the support varies between executions, often doing no better than importance sampling from the prior. To address this, we introduce a new inference framework: Divide, Conquer, and Combine, which remains efficient for such models, and show how it can be implemented as an automated and generic PPS inference engine. We empirically demonstrate substantial performance improvements over existing approaches on three examples.