Comparing Predictive Inference Methods for Discrete Domains

Petri Kontkanen, Petri Myllymäki, Tomi Silander, Henry Tirri, Peter Grünwald
Proceedings of the Sixth International Workshop on Artificial Intelligence and Statistics, PMLR R1:311-318, 1997.

Abstract

Predictive inference is seen here as the process of determining the predictive distribution of a discrete variable, given a data set of training examples and the values for the other problem domain variables. We consider three approaches for computing this predictive distribution, and assume that the joint probability distribution for the variables belongs to a set of distributions determined by a set of parametric models. In the simplest case, the predictive distribution is computed by using the model with the \emph{maximum a posteriori (MAP)} posterior probability. In the \emph{evidence} approach, the predictive distribution is obtained by averaging over all the individual models.in the model family. In the third case, we define the predictive distribution by using Rissanen’s new definition of \emph{stochastic complexity}. Our experiments performed with the family of Naive Bayes models suggest that when using all the data available, the stochastic complexity approach produces the most accurate predictions in the log-score sense. However, when the amount of available training data is decreased, the evidence approach clearly outperforms the two other approaches. The MAP predictive distribution is clearly inferior in the log-score sense to the two more sophisticated approaches, but for the 0/1-score the MAP approach may still in some cases produce the best results.

Cite this Paper


BibTeX
@InProceedings{pmlr-vR1-kontkanen97a, title = {Comparing Predictive Inference Methods for Discrete Domains}, author = {Kontkanen, Petri and Myllym\"aki, Petri and Silander, Tomi and Tirri, Henry and Gr\"unwald, Peter}, booktitle = {Proceedings of the Sixth International Workshop on Artificial Intelligence and Statistics}, pages = {311--318}, year = {1997}, editor = {Madigan, David and Smyth, Padhraic}, volume = {R1}, series = {Proceedings of Machine Learning Research}, month = {04--07 Jan}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/r1/kontkanen97a/kontkanen97a.pdf}, url = {https://proceedings.mlr.press/r1/kontkanen97a.html}, abstract = {Predictive inference is seen here as the process of determining the predictive distribution of a discrete variable, given a data set of training examples and the values for the other problem domain variables. We consider three approaches for computing this predictive distribution, and assume that the joint probability distribution for the variables belongs to a set of distributions determined by a set of parametric models. In the simplest case, the predictive distribution is computed by using the model with the \emph{maximum a posteriori (MAP)} posterior probability. In the \emph{evidence} approach, the predictive distribution is obtained by averaging over all the individual models.in the model family. In the third case, we define the predictive distribution by using Rissanen’s new definition of \emph{stochastic complexity}. Our experiments performed with the family of Naive Bayes models suggest that when using all the data available, the stochastic complexity approach produces the most accurate predictions in the log-score sense. However, when the amount of available training data is decreased, the evidence approach clearly outperforms the two other approaches. The MAP predictive distribution is clearly inferior in the log-score sense to the two more sophisticated approaches, but for the 0/1-score the MAP approach may still in some cases produce the best results.}, note = {Reissued by PMLR on 30 March 2021.} }
Endnote
%0 Conference Paper %T Comparing Predictive Inference Methods for Discrete Domains %A Petri Kontkanen %A Petri Myllymäki %A Tomi Silander %A Henry Tirri %A Peter Grünwald %B Proceedings of the Sixth International Workshop on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 1997 %E David Madigan %E Padhraic Smyth %F pmlr-vR1-kontkanen97a %I PMLR %P 311--318 %U https://proceedings.mlr.press/r1/kontkanen97a.html %V R1 %X Predictive inference is seen here as the process of determining the predictive distribution of a discrete variable, given a data set of training examples and the values for the other problem domain variables. We consider three approaches for computing this predictive distribution, and assume that the joint probability distribution for the variables belongs to a set of distributions determined by a set of parametric models. In the simplest case, the predictive distribution is computed by using the model with the \emph{maximum a posteriori (MAP)} posterior probability. In the \emph{evidence} approach, the predictive distribution is obtained by averaging over all the individual models.in the model family. In the third case, we define the predictive distribution by using Rissanen’s new definition of \emph{stochastic complexity}. Our experiments performed with the family of Naive Bayes models suggest that when using all the data available, the stochastic complexity approach produces the most accurate predictions in the log-score sense. However, when the amount of available training data is decreased, the evidence approach clearly outperforms the two other approaches. The MAP predictive distribution is clearly inferior in the log-score sense to the two more sophisticated approaches, but for the 0/1-score the MAP approach may still in some cases produce the best results. %Z Reissued by PMLR on 30 March 2021.
APA
Kontkanen, P., Myllymäki, P., Silander, T., Tirri, H. & Grünwald, P.. (1997). Comparing Predictive Inference Methods for Discrete Domains. Proceedings of the Sixth International Workshop on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research R1:311-318 Available from https://proceedings.mlr.press/r1/kontkanen97a.html. Reissued by PMLR on 30 March 2021.

Related Material