Diagnostic Measures for Model Criticism

By C. Carota, Giovanni Parmigiani, Nicholas G. Polson. 1996

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Kacper P. Chwialkowski
2. Wittawat Jitkrittum
3. Krikamol Muandet
4. Arthur Gretton
5. Zoltán Szabó
6. David Lopez-Paz
7. Kenji Fukumizu
8. Bharath K. Sriperumbudur

Citations in the corpus

• Interpretable Distribution Features with Maximum Testing Power (2016, methodology): “…measures is to model criticism, where samples from a trained model are compared with a validation sample: in the univariate case, through the KL divergence (Cinzia Carota and Polson, 1996), or in the multivariate case, by use of the maximum mean discrepancy (MMD) (Lloyd and Ghahramani, 2015).”
• Approximate Models and Robust Decisions (2014, methodology): “Also of note is the idea of model elaboration (Carota, Parmigiani and Polson, 1996), whereby a parametric model is embedded into a larger class of models (e.g., a normal model into the family of student-t distributions).”