Ye-Yi Wang and Alex Acero
Maximum entropy (MaxEnt) models have been used in many spoken language tasks. The training of a MaxEnt model often involves an iterative procedure that starts from an initial parameterization and gradually updates it towards the optimum. Due to the convexity of its objective function (hence a global optimum on a training set), little attention has been paid to model initialization in MaxEnt training. However, MaxEnt model training often ends early before convergence to the global optimum, and prior distributions with hyper-parameters are often added to the objective function to prevent over-fitting. This paper shows that the initialization and regularization hyper-parameter setting may significantly affect the test set accuracy. It investigates the MaxEnt initialization/regularization based on an n-gram classifier and a TF*IDF weighted vector space model. The theoretically motivated TF*IDF initialization/regularization has achieved significant improvements over the baseline flat initialization/regularization. In contrast, the n-gram based initialization/regularization often does not exhibit significant improvements.
In IEEE Automatic Speech Recognition and Understanding Workshop
Publisher Institute of Electrical and Electronics Engineers, Inc.
© 2007 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.