Sökning: "Lawrence Murray"

Hittade 3 avhandlingar innehållade orden Lawrence Murray.

  1. 1. Probabilistic Programming for Birth-Death Models of Evolution

    Författare :Jan Kudlicka; Johannes Borgström; Thomas B. Schön; Lawrence M. Murray; Bret Larget; Uppsala universitet; []
    Nyckelord :NATURVETENSKAP; NATURAL SCIENCES; probabilistic programming; birth-death models; statistical phylogenetics; particle filters; sequential Monte Carlo SMC ; Computer Science; Datavetenskap;

    Sammanfattning : Phylogenetic birth-death models constitute a family of generative models of evolution. In these models an evolutionary process starts with a single species at a certain time in the past, and the speciations—splitting one species into two descendant species—and extinctions are modeled as events of non-homogenous Poisson processes. LÄS MER

  3. 2. Correct and Efficient Monte Carlo Inference for Universal Probabilistic Programming Languages

    Författare :Daniel Lundén; David Broman; Lawrence Murray; Joakim Jaldén; Sam Staton; KTH; []
    Nyckelord :NATURVETENSKAP; NATURAL SCIENCES; Probabilistic programming languages; Compilers; Static program analysis; Monte Carlo inference; Operational semantics; Probabilistiska programmeringsspråk; Kompilatorer; Statisk programanalys; Monte Carlo-inferens; Operationell semantik; Informations- och kommunikationsteknik; Information and Communication Technology;

    Sammanfattning : Probabilistic programming languages (PPLs) allow users to express statistical inference problems that the PPL implementation then, ideally, solves automatically. In particular, PPL users can focus on encoding their inference problems, and need not concern themselves with the intricacies of inference. LÄS MER

  4. 3. Sequential Monte Carlo methods for conjugate state-space models

    Författare :Anna Wigren; Fredrik Lindsten; Lawrence Murray; Riccardo Sven Risuleo; Simon Maskell; Uppsala universitet; []
    Nyckelord :TEKNIK OCH TEKNOLOGIER; ENGINEERING AND TECHNOLOGY; Sequential Monte Carlo; Particle filter; Markov chain Monte Carlo; Conjugacy; State-space model; Probabilistic programming; Electrical Engineering with specialization in Signal Processing; Elektroteknik med inriktning mot signalbehandling;

    Sammanfattning : Bayesian inference in state-space models requires the solution of high-dimensional integrals, which is intractable in general. A viable alternative is to use sample-based methods, like sequential Monte Carlo, but this introduces variance into the inferred quantities that can sometimes render the estimates useless. LÄS MER