
Learning Probabilistic Programs
We develop a technique for generalising from data in which models are sa...
read it

Particle Gibbs with Ancestor Sampling for Probabilistic Programs
Particle Markov chain Monte Carlo techniques rank among current stateof...
read it

Engineering a Fast Probabilistic Isomorphism Test
We engineer a new probabilistic MonteCarlo algorithm for isomorphism te...
read it

Deriving Probability Density Functions from Probabilistic Functional Programs
The probability density function of a probability distribution is a fund...
read it

OutputSensitive Adaptive MetropolisHastings for Probabilistic Programs
We introduce an adaptive outputsensitive MetropolisHastings algorithm ...
read it

CoarsetoFine Sequential Monte Carlo for Probabilistic Programs
Many practical techniques for probabilistic inference require a sequence...
read it

Hybrid Probabilistic Programs: Algorithms and Complexity
Hybrid Probabilistic Programs (HPPs) are logic programs that allow the p...
read it
Compositional Inference Metaprogramming with Convergence Guarantees
Inference metaprogramming enables effective probabilistic programming by supporting the decomposition of executions of probabilistic programs into subproblems and the deployment of hybrid probabilistic inference algorithms that apply different probabilistic inference algorithms to different subproblems. We introduce the concept of independent subproblem inference (as opposed to entangled subproblem inference in which the subproblem inference algorithm operates over the full program trace) and present a mathematical framework for studying convergence properties of hybrid inference algorithms that apply different MarkovChain Monte Carlo algorithms to different parts of the inference problem. We then use this formalism to prove asymptotic convergence results for probablistic programs with inference metaprogramming. To the best of our knowledge this is the first asymptotic convergence result for hybrid probabilistic inference algorithms defined by (subproblembased) inference metaprogramming.
READ FULL TEXT
Comments
There are no comments yet.