Information

Check out the new study results page. Another step closer to all the right info in one place.

What is your study programme?
Talk

Presentation Marjolein Fokkema on Friday 18 november 2022

Last modified on 10-11-2022
Student of the College and Graduate School of Child Development and Education receive 1 colloquium point for attending the entire presentation.
Show information for your study programme
You're currently viewing general information. Choose your study programme to see additional information that's specific to your study programme, such as deadlines, regulations and contact details.
What is your study programme?
Start date
18-11-2022 14:00
End date
18-11-2022 15:00
Location

REC-C2.04

Presentation Marjolein Fokkema: 'Prediction rule ensembles: Balancing interpretability and accuracy in statistical prediction'

Prediction Rule Ensembles (PREs) aim to strike a balance between the high predictive accuracy of black-box decision tree ensembles (e.g., random forests, boosted tree ensembles), and the ease of interpretability of sparse regression methods and single decision trees. 

PREs have been shown to provide predictive accuracy competitive with random forests, but consist of a small number of rules and thus may be easier to interpret for human decision makers. As such, PREs may contribute to bridging the gap between clinical research and practice in psychology. 

For interpretable and explainable AI, PRE may offer a two-fold advantage: First, it provides an inherently interpretable prediction method. Second, it can be used for explanation of black-box methods. In the current presentation, I would like to present applications to psychological research data, and discuss initial results on causal modeling.

Colloquiumcategorie: Lezingen en congressen