A Gaussian prior for smoothing maximum entropy models

DMCA

A Gaussian prior for smoothing maximum entropy models (1999)

Cached

Download Links

by Stanley F. Chen , Ronald Rosenfeld

Citations:

251 - 2 self

BibTeX

@TECHREPORT{Chen99agaussian,
    author = {Stanley F. Chen and Ronald Rosenfeld},
    title = {A Gaussian prior for smoothing maximum entropy models},
    institution = {},
    year = {1999}
}

OpenURL

Abstract

In certain contexts, maximum entropy (ME) modeling can be viewed as maximum likelihood train-ing for exponential models, and like other maximum likelihood methods is prone to overfitting of training data. Several smoothing methods for maximum entropy models have been proposed to address this problem, but previous results do not make it clear how these smoothing methods com-pare with smoothing methods for other types of related models. In this work, we survey previous work in maximum entropy smoothing and compare the performance of several of these algorithms with conventional techniques for smoothing n-gram language models. Because of the mature body of research in n-gram model smoothing and the close connection between maximum entropy and conventional n-gram models, this domain is well-suited to gauge the performance of maximum entropy smoothing methods. Over a large number of data sets, we find that an ME smoothing method proposed to us by Lafferty [1] performs as well as or better than all other algorithms under consideration. This general and efficient method involves using a Gaussian prior on the parame-ters of the model and selecting maximum a posteriori instead of maximum likelihood parameter values. We contrast this method with previous n-gram smoothing methods to explain its superior performance.

Keyphrases

maximum entropy model gaussian prior maximum entropy conventional technique n-gram language model large number maximum entropy smoothing maximum likelihood parameter value previous result efficient method exponential model superior performance survey previous work n-gram model smoothing maximum likelihood method conventional n-gram model mature body close connection maximum likelihood train-ing certain context related model data set

DMCA