2000 character limit reached
Automatic Rao-Blackwellization for Sequential Monte Carlo with Belief Propagation
Published 15 Dec 2023 in cs.LG and stat.CO | (2312.09860v1)
Abstract: Exact Bayesian inference on state-space models~(SSM) is in general untractable, and unfortunately, basic Sequential Monte Carlo~(SMC) methods do not yield correct approximations for complex models. In this paper, we propose a mixed inference algorithm that computes closed-form solutions using belief propagation as much as possible, and falls back to sampling-based SMC methods when exact computations fail. This algorithm thus implements automatic Rao-Blackwellization and is even exact for Gaussian tree models.
- Semi-symbolic inference for efficient streaming probabilistic programming. In OOPSLA, 2022.
- Reactive probabilistic programming. In PLDI. ACM, 2020.
- The synchronous languages 12 years later. Proc. IEEE, 91(1):64–83, 2003.
- Pyro: Deep universal probabilistic programming. J. Mach. Learn. Res., 20:28:1–28:6, 2019.
- An introduction to sequential Monte Carlo. Springer, 2020.
- Gen: A general-purpose probabilistic programming system with programmable inference. In Proceedings of the 40th ACM SIGPLAN Conference on Programming Language Design and Implementation, PLDI 2019, pp. 221–236.
- A tutorial on particle filtering and smoothing: Fifteen years later. Handbook of nonlinear filtering, 12(656-704):3, 2009.
- Turing: a language for flexible probabilistic inference. In AISTATS 2018.
- The design and implementation of probabilistic programming languages, 2014. URL http://dippl.org. Accessed April 2020.
- Novel approach to nonlinear/non-gaussian bayesian state estimation. In IEE proceedings F (radar and signal processing), volume 140, pp. 107–113. IET, 1993.
- A differentiable programming system to bridge machine learning and scientific computing. CoRR, abs/1907.07587, 2019.
- Automated learning with a probabilistic programming language: Birch. Annual Reviews in Control, 46:29–43, 2018.
- Delayed sampling and automatic rao-blackwellization of probabilistic programs. In AISTATS, 2018.
- Design and implementation of probabilistic programming language anglican. In IFL, pp. 6:1–6:12. ACM, 2016.
- Correctness of belief propagation in gaussian graphical models of arbitrary topology. Advances in neural information processing systems, 12, 1999.
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Top Community Prompts
Collections
Sign up for free to add this paper to one or more collections.