Inference Plans for Hybrid Particle Filtering

TL;DR

This paper introduces inference plans, a programming interface for hybrid particle filtering in probabilistic programming languages, enabling developers to control variable partitioning, resulting in significant speed and accuracy improvements validated through static analysis and benchmarks.

Contribution

It presents inference plans for better control over variable partitioning in hybrid particle filtering and introduces Siren, a PPL supporting annotation-based inference plan specification and static analysis.

Findings

Speed ups of 1.76x on average and up to 206x for target accuracy.

Accuracy improvements of 1.83x on average and up to 595x.

Static analysis identifies all satisfiable plans in 27 of 33 cases.

Abstract

Advanced probabilistic programming languages (PPLs) using hybrid particle filtering combine symbolic exact inference and Monte Carlo methods to improve inference performance. These systems use heuristics to partition random variables within the program into variables that are encoded symbolically and variables that are encoded with sampled values, and the heuristics are not necessarily aligned with the developer's performance evaluation metrics. In this work, we present inference plans, a programming interface that enables developers to control the partitioning of random variables during hybrid particle filtering. We further present Siren, a new PPL that enables developers to use annotations to specify inference plans the inference system must implement. To assist developers with statically reasoning about whether an inference plan can be implemented, we present an abstract-interpretation-based static analysis for Siren for determining inference plan satisfiability. We prove the analysis is sound with respect to Siren's semantics. Our evaluation applies inference plans to three different hybrid particle filtering algorithms on a suite of benchmarks. It shows that the control provided by inference plans enables speed ups of 1.76x on average and up to 206x to reach a target accuracy, compared to the inference plans implemented by default heuristics; the results also show that inference plans improve accuracy by 1.83x on average and up to 595x with less or equal runtime, compared to the default inference plans. We further show that our static analysis is precise in practice, identifying all satisfiable inference plans in 27 out of the 33 benchmark-algorithm evaluation settings.