Learn 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)
SUMMARY We propose an approach to constructing a new type of design, a sliced space-filling design, intended for computer experiments with qualitative and quantitative factors. The approach starts with constructing a Latin hypercube design based on a special orthogonal array for the quantitative factors and then partitions the design into groups(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)
We give an optimal dynamic programming algorithm to solve a class of finite-horizon decentralized Markov decision processes (MDPs). We consider problems with a broadcast information structure that consists of a central node that only has access to its own state but can affect several outer nodes, while each outer node has access to both its own state and(More)
UNLABELLED The treatment of 300-mg/day isoflavones (aglycone equivalents) (172.5 mg genistein + 127.5 mg daidzein) for 2 years failed to prevent lumbar spine and total proximal femur bone mineral density (BMD) from declining as compared with the placebo group in a randomized, double-blind, two-arm designed study enrolling 431 postmenopausal women 45-65(More)
To my parents and my wife Ruoyan iii ACKNOWLEDGEMENTS Foremost I would like to express my deep gratitude to my advisor, Professor C. F. Jeff Wu. His inspiration, guidance, encouragement and insight helped me through these valuable years at Georgia Tech and the University of Michigan. I am grateful to Dr. Yasuo Amemiya, my co-advisor, for his guidance,(More)
Church is a Turing-complete probabilistic programming language, designed for inference. By allowing for easy description and manipulation of distributions, it allows one to describe classical Al models in compact ways, providing a language for very rich expression. However, for inference in Bayes nets, Hidden Markov Models, and topic models, the very(More)