Bayesian State-Space Modelling on High-Performance Hardware Using LibBi

By Lawrence M. Murray. 2013

116 citations recorded by the source corpus.

LibBi is a software package for state space modelling and Bayesian inference on modern computer hardware, including multi-core central processing units, many-core graphics processing units, and distributed-memory clusters of such devices. The software parses a domain-specific language for model specification, then optimizes, generates, compiles and runs code for the given model, inference method and hardware platform. In presenting the software, this work serves as an introduction to state space models and the specialized methods developed for Bayesian inference with them. The focus is on sequential Monte Carlo (SMC) methods such as the particle filter for state estimation, and the particle Markov chain Monte Carlo and SMC2 methods for parameter estimation. All are well-suited to current computer hardware. Two examples are given and developed throughout, one a linear three-element windkessel model of the human arterial system, the other a nonlinear Lorenz '96 model. These are specified in the prescribed modelling language, and LibBi demonstrated by performing inference with them. Empirical results are presented, including a performance comparison of the software with different hardware configurations.

Subfields: Artificial Intelligence, Control and Systems Engineering, Probabilistic & Bayesian Methods, ML Systems & Hardware

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Control and Systems Engineering

In Probabilistic & Bayesian Methods

In ML Systems & Hardware

Citations in the corpus

An Introduction to Probabilistic Programming (2018, neutral): “, 2014), LibBi (Murray, 2013), Birch (Murray et al.”
MXFusion: A Modular Deep Probabilistic Programming Library (2018, background): “Probabilistic programming language (PPL, Lunn et al., 2000; Pfeffer, 2001; Goodman et al., 2008; Pfeffer, 2009; Murray, 2013; Paige & Wood, 2014; Minka et al., 2014; Carpenter et al., 2017; Ge et al., 2018) is a similar domain specific programming language that aims at describing probabilistic…”
Probabilistic programming with programmable inference (2018, methodology): “For example, LibBi [34] and Gen [27] support user-space definition of custom proposal distributions, and Figaro [35], Edward [49], and Pyro [1] support additional constructs.”
Turing: A Language for Flexible Probabilistic Inference (2018, background): “Probabilistic programming languages [Goodman et al., 2008, Wood et al., 2014, Lunn et al., 2000, Minka et al., 2014, Stan Development Team, 2014, Murray, 2013, Pfeffer, 2001, 2009, Paige and Wood, 2014, Murray et al., 2017] aim to fill this gap by providing a very flexible framework for defining…”
Sequential Monte Carlo with Highly Informative Observations (2014, methodology): “Experiments are conducted using the LibBi software [33], in which the methods have been implemented under the name bridge particle filter.”
Delayed Sampling and Automatic Rao-Blackwellization of Probabilistic Programs (2017, methodology): “…Carlo (SMC) inference algorithms [4], extending an arc of research that includes probabilistic programming languages (PPLs) such as Venture [14], Anglican [26], Probabilistic C [18], WebPPL [8], Fi-garo [20], and Turing [7], as well as similarly-motivated software such as LibBi [16] and BiiPS [25].”
Gen: a general-purpose probabilistic programming system with programmable inference (2019, background): “Some languages are designed to be well-suited for a specific domain, such as hierarchical Bayesian statistics (Stan [6]), deep generative modeling (Pyro [5]), or uncertainty quantification for scientific simulations (LibBi [33]).”
Biips: Software for Bayesian Inference with Interacting Particle Systems (2014, methodology): “LibBi (Murray 2013) is another similar software that implements SMC methods and is suited to parallel and distributed computer hardware such as multi-core CPUs, GPUs and clusters.”