Help Guide Disclaimer Contact us Login
  Advanced SearchBrowse





Introduction to Bayesian Experimental Design


Tanner,  T
Department Human Perception, Cognition and Action, Max Planck Institute for Biological Cybernetics, Max Planck Society;

There are no locators available
Fulltext (public)
There are no public fulltexts available
Supplementary Material (public)
There is no public supplementary material available

Tanner, T. (2005). Introduction to Bayesian Experimental Design. Talk presented at 6. Neurowissenschaftliche Nachwuchskonferenz Tübingen (NeNa '05). Blaubeuren, Germany.

Cite as:
Scientists perform experiments to collect evidence supporting one or another hypothesis or theory. Experimentation requires decisions about how an experiment will be conducted and analyzed, such as the necessary sample size, selection of treatments or choice of statistical tests. These decisions depend on the goals and purpose of the experiment and may be constrained by available resources and ethical considerations. Prior knowledge is usually available from previous experiments or existing theories motivating the investigation. The Bayesian approach provides a coherent framework for combining prior information, theoretical models and uncertainties regarding unknown quantities to find an experimental design optimizing the goals of the investigation. Applying methods of Bayesian experimental design may help increasing the efficiency and the informativeness of an experiment. In this talk I will give an short introduction to Bayesian inference and decision theory, followed by an overview over Bayesian experimental design and one example of how to design an Bayesian optimal experiment.