Skip to content

fhuszar/thesis

Repository files navigation

thesis

Ferenc Huszar's PhD thesis on scoring rules and Bayesian inference.

If you like my open-source thesis, you can donate to my after-thesis-submission-chill fund via micropayments service Flattr: Flattr this git repo

Contents:

  • General intro to scoring rules
  • Visualisation of information geometries induced by scoring rules
  • loss-calibrated approximate Bayesian inference
  • A scoring rule-based framework Bayesian active learning
  • Bayesian active learning for GP classification and preference learning
  • Adaptive Bayesian Quantum Tomography

## Licenses and stuff:

  • Most stuff, including figures are up for grabs, with appropriate reference to the thesis
  • The Cambridge University logo is from http://www.admin.cam.ac.uk/offices/communications/services/logos/ and may only be used as advised on that website
  • The repo contains latex class files or matlab macros written by other people, usually the file contains the name of the authors

About

my PhD thesis on Bayesian inference

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published