Kammar, Ohad

Goldstein, Oliver

2024-12-20T14:52:25Z

2024-12-20T14:52:25Z

2019-11-23

Probabilistic programming languages, which exist in abundance, are languages that allow users to calculate probability distributions defined by probabilistic programs, by using inference algorithms. However, the underlying inference algorithms are not implemented in a modular fashion, though, the algorithms are presented as a composition of other inference components. This discordance between the theory and the practice of Bayesian machine learning, means that reasoning about the correctness of probabilistic programs is more difficult, and composing inference algorithms together in code may not necessarily produce correct compound inference algorithms. In this dissertation, I create a modular probabilistic programming library, already a nice property as its not a standalone language, called Koka Bayes, that is based off of both the modular design of Monad Bayes – a probabilistic programming library developed in Haskell – and its semantic validation. The library is embedded in a recently created programming language, Koka, that supports algebraic effect handlers and expressive effect types – novel programming abstractions that support modular programming. Effects are generalizations of computational side-effects, and it turns out that fundamental operations in probabilistic programming such as probabilistic choice and conditioning are instances of effects.

https://hdl.handle.net/1842/42934

http://dx.doi.org/10.7488/era/5485

en

The University of Edinburgh

effect handlers

algebraic effects

Bayesian inference

probabilistic programming

statistical modelling

climate change modelling

koka

generative modelling

importance sampling

Markov Chain Monte Carl

resample-move sequential Monte Carlo

Particle Marginal Metropolis Hastings

Kalman Filters

Berkeley Earth Dataset

hyper-parameter inference

Modular probabilistic programming with algebraic effects

Thesis or Dissertation

Masters

MSc Master of Science