Models and Algorithms for Sequential Decision Problems Under Uncertainty (19w5231)
The Banff International Research Station will host the "Models and Algorithms for Sequential Decision Problems under Uncertainty" workshop from January 13th to January 18th, 2019. In most real world problems including engineering system design, risk management in financial and other domains, pricing and revenue management and many other applications, critical decisions are often made sequentially and in the face of uncertainties. These uncertainties arise from several factors including statistical errors in parameters estimation, uncertainty in measurement, uncertainty in future exogenous variables and uncertainty in model correctness itself. Therefore, sequential decision problems under uncertainty are an important class of problems both from a theoretical as well as practical point of view, relevant to many areas of applied science, including statistical estimation and inference, and control theory, and have been extensively studied in the literature. Various approaches have developed to model and solve these sequential decision problems; prominent among these being Stochastic Optimization, Robust Optimization and Online Optimization approaches. While the Operations Research community has focus extensively on robust and stochastic optimization approaches, the online optimization approach has been mainly studied in the Computer Science community. This workshop will bring together leading researchers both senior and young from different fields including Operations Research, Statistics, and Computer Science to focus on these different paradigms for sequential decision problems under uncertainty. The workshop will aim to provide an overview of the state-of-the-art in these different fields; with a goal to stimulate discussions and exchange of ideas between researchers from these different fields that typically would not get a chance to interact and collaborate. The focus will be to explore relationships between these different approaches, more specifically, between robust optimization and online optimization.
