Bayesian mixed-effects inference on classification performance in hierarchical datasets
Lists
Only you see this
Abstract
• Bayesian mixed-effects inference for group studies (i) models within-subject and across-subjects uncertainty, (ii) enables conclusions in terms of intuitive probability statements, and (iii) can be used with various performance measures, e.g., the balanced accuracy. In contrast to previous sampling algorithms, our variational approximation is computationally highly efficient. • We hope that our approach will improve the precision and interpretability of
BibTeX
@inproceedings{brodersen2012bayesian,
author = {Kay H. Brodersen and Christoph Mathys and Justin Chumbley and Jean Daunizeau and Cheng Soon Ong and Joachim M. Buhmann and Klaas Ε. Stephan},
title = {{Bayesian mixed-effects inference on classification performance in hierarchical datasets}},
booktitle = {Zurich},
year = {2012},
doi = {10.5167/uzh-71594},
}Loading related content...