Learn More
Absfract-We propose, implement, and evaluate a class of nonstationary-state hidden Markov models (HMM's) having each state associated with a distinct polynomial regression function of time plus white Gaussian noise. The model represents the transitional acoustic trajectories of speech in a parametric manner, and includes the standard stationary-state HMM as(More)
A class of sequential designs for estimating the percentiles of a quantal response curve is proposed. Its updating rule is based on an efficient sumary of all the data available via a parametric model. The "logit-MLEV version of the proposed designs can be viewed as a natural analogue of the Robbins-Monro procedure in the case of binary data. It is shown to(More)
Preliminary design of a complex system often involves exploring a broad design space. This may require repeated use of computationally expensive simulations. To ease the computational burden, surrogate models are built to provide rapid approximations of more expensive models. However, the surrogate models themselves are often expensive to build because they(More)
A smooth response surface (SRS) algorithm is developed as an elaborate data mining technique for analyzing gene expression data and constructing a gene regulatory network. A three-dimensional SRS is generated to capture the biological relationship between the target and activator-repressor. This new technique is applied to functionally describe triplets of(More)
By studying treatment contrasts and ANOVA models, we propose a generalized minimum aberration criterion for comparing asymmetrical fractional factorial designs. The criterion is independent of the choice of treatment contrasts and thus model-free. It works for symmetrical and asymmetrical designs, regular and nonregular designs. In particular, it reduces to(More)
Modeling experiments with qualitative and quantitative factors is an important issue in computer modeling. A framework for building Gaussian process models that incorporate both types of factors is proposed. The key to the development of these new models is an approach for constructing correlation functions with qualitative and quantitative factors. An(More)
Standard practice in analyzing data from different types of experiments is to treat data from each type separately. By borrowing strength across multiple sources, an integrated analysis can produce better results. Careful adjustments need to be made to incorporate the systematic differences among various experiments. To this end, some Bayesian hierarchical(More)
Categorical data arise quite often in industrial experiments because of an expensive or inadequate measurement system for obtaining continuous data. When the failure probability/defect rate is small, experiments with categorical data provide little information regarding the effect of factors of interests and are generally not useful for product/process(More)