US-Poland Computation Research on Decomposition Methods for Linear Multistage Stocastic Programming

这个由普林斯顿大学的John Mulvey博士和华沙理工大学的Andrzey Ruszczynski博士合作的美国-波兰研究项目的主要目标是研究解决具有动态结构和不确定参数的大规模优化问题的计算方法。研究人员打算把重点放在分解类型的方法上，这种方法将问题分解成与时间阶段或场景相对应的更小的部分。他们计划采用以下方法：1)应用正则化和增广拉格朗日的原始和对偶分解；2)特殊版本的势垒或内点法。本课题在线性规划中的研究成果有望为解决大型多阶段随机规划问题开辟新的可能性。这个计算机科学研究项目通过使美国和东欧的专家在共同感兴趣和能力强的领域结合互补的人才和集中研究资源，实现了推进科学知识的计划目标。

The primary objective of this US-Poland research project between Dr. John Mulvey of Princeton University and Dr. Andrzey Ruszczynski of Warsaw University of Technology is to study computational methods for solving large-scale optimization problems that have a dynamic structure and uncertain parameters. The researchers intend to focus on decomposition-type methods which split problems into smaller parts corresponding to time stages or scenarios. They plan to employ the following methods: 1) primal and dual decomposition with the application of regularization and augmented Lagrangians and 2) special versions of barrier or interior point methods. The results of this project in linear programming are expected to open new possibilities for solving large-multistage stochastic programming problems. This project in computer science research fulfills the program objective of advancing scientific knowledge by enabling experts in the United States and Eastern Europe to combine complementary talents and pool research resources in areas of strong mutual interest and competence.