Sampling First Order Logical Particles
This work addresses the problem of high-precision state estimation for applications like diagnosis and robotics, presenting an incremental improvement over existing sampling methods.
The paper tackles approximate inference in dynamic systems by introducing an algorithm that samples deterministic executions using first-order logic representations, resulting in a smaller expected error compared to propositional and Sequential Monte Carlo sampling techniques.
Approximate inference in dynamic systems is the problem of estimating the state of the system given a sequence of actions and partial observations. High precision estimation is fundamental in many applications like diagnosis, natural language processing, tracking, planning, and robotics. In this paper we present an algorithm that samples possible deterministic executions of a probabilistic sequence. The algorithm takes advantage of a compact representation (using first order logic) for actions and world states to improve the precision of its estimation. Theoretical and empirical results show that the algorithm's expected error is smaller than propositional sampling and Sequential Monte Carlo (SMC) sampling techniques.