Learning Graphical Model Parameters with Approximate Marginal Inference
Author:
- 25 January 2013
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 35 (10) , 2454-2467
Abstract
Likelihood-based learning of graphical models faces challenges of computational complexity and robustness to model misspecification. This paper studies methods that fit parameters directly to maximize a measure of the accuracy of predicted marginals, taking into account both model and inference approximations at training time. Experiments on imaging problems suggest marginalization-based learning performs better than likelihood-based approximations on difficult problems where the model being fit is approximate in nature.Keywords
All Related Versions
This publication has 36 references indexed in Scilit:
- Structured Learning and Prediction in Computer VisionFoundations and Trends in Computer Graphics and Vision, 2011
- Accelerated conjugate gradient algorithm with finite difference Hessian/vector product approximation for unconstrained optimizationJournal of Computational and Applied Mathematics, 2009
- Fields of ExpertsInternational Journal of Computer Vision, 2009
- Measuring uncertainty in graph cut solutionsComputer Vision and Image Understanding, 2008
- TextonBoost for Image Understanding: Multi-Class Object Recognition and Segmentation by Jointly Modeling Texture, Layout, and ContextInternational Journal of Computer Vision, 2007
- Robust model-based scene interpretation by multilayered context informationComputer Vision and Image Understanding, 2007
- Model distortions in Bayesian MAP reconstructionInverse Problems and Imaging, 2007
- Learning in Gibbsian fields: how accurate and how fast can it be?IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002
- Composite likelihood methods1988
- Probabilistic Solution of Ill-Posed Problems in Computational VisionJournal of the American Statistical Association, 1987