ML Seminar: Rianne de Heide (CWI/Leiden University)

Everyone is welcome to attend the ML seminar of Rianne de Heide with the title 'The truth-convergence of open-minded Bayesianism'.
  • What English Machine Learning Seminars
  • When 14-06-2019 from 11:00 to 12:00 (Europe/Amsterdam / UTC200)
  • Where Room L120 at CWI, Science Park 123 in Amsterdam
  • Web Visit external website
  • Add event to calendar iCal

Everyone is welcome to attend the ML seminar of Rianne de Heide with the title 'The truth-convergence of open-minded Bayesianism'.

Joint work with Tom F. Sterkenburg

Submitted; Accepted to the Formal Epistemology Workshop 2019, and the biennial conference of the European Philosophy of Science Association.
Wenmackers and Romeijn (2016) formalize ideas going back to Shimony (1970) and Putnam (1963) into an open-minded Bayesian inductive logic, that can dynamically incorporate hypotheses that are proposed in the course of the learning process. In this paper, we show that Wenmackers and Romeijn’s proposal does not preserve the classical Bayesian consistency guarantee of almost-sure merger with the true hypothesis. We diagnose the problem, and offer two versions of forward-looking open-minded Bayesians that, for any given scheme for the generation of hypotheses and prior/posterior probabilities, do preserve this guarantee.

Main reference: S. Wenmackers and J.-W. Romeijn. New theory about old evidence: A framework for open-minded Bayesianism. Synthese, 193(4):1225–1250, 2016.