Skip to content

Comparing expectation-based updating of eight Bayesian source-reliability models.

License

Notifications You must be signed in to change notification settings

klee-schoeppl/source-reliability

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 

Repository files navigation

Interpreting and evaluating Bayesian source-reliability models

This code, written for my Master's thesis at the MCMP, puts eight agents to the task of determining the reliability of an advisor via expectation-based updating on multiple rounds of testimony. It is similar to the model used by Hahn et al. (2018) & Merdes et al. (2021) in that it places these focal agents in a situation without any direct access to actual outcomes. Inspired by Duijf (2021), my model disentangles reliability into comptency and alignment and evaluates agents in terms of their trustworthiness-accuracies in addition to their (change in) veritistic values (Goldman 1999). The Boolean BH-model was introduced Bovens & Hartmann (2004), the continuous OL-model by Olsson (2011) & Angere (2010), the Boolean AL-model by Assaad (2022) & Harris et al. (2016), and in creating the SD-model I was again inspired by Duijf (2021).

References:

  • Adam Harris, Ulrike Hahn, Jens Madsen & Anne Hsu (2016) “The appeal to expert opinion: Quantitative support for a Bayesian network approach.”
  • Erik J. Olsson (2011) “A simulation approach to veritistic social epistemology.”
  • Sofie Angere (2010) “Knowledge in a social network.”
  • Leon Assaad (2022) “Trust in Information Sources: Reconstructing Worrisome Phenomena Using Bayesian Models of Source Reliability.”
  • Luc Bovens & Stephan Hartmann (2004) “Bayesian epistemology.”
  • Hein Duijf (2021) “Should one trust experts?”
  • Ulrike Hahn, Christoph Merdes & Momme von Sydow (2018) “How good is your evidence and how would you know?”
  • Christoph Merdes, Momme von Sydow & Ulrike Hahn (2021) “Formal models of source reliability.”
  • Alvin Goldman (1999) “Knowledge in a social world.”

About

Comparing expectation-based updating of eight Bayesian source-reliability models.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published