This shows you the differences between two versions of the page.
dynamic_learning_and_information_information_discovery-_an_adaptive_optimization_perspective [2016/09/01 19:15] (current)
|Line 1:||Line 1:|
|+||Stochastic programming and robust optimization are disciplines concerned with optimal decision-making under uncertainty over time. Traditional models and solution algorithms have been tailored to problems where the controller cannot influence the realization of the uncertain parameters nor the timing of their revelation. Nevertheless, these assumptions fail to hold in numerous real-world applications.|
|+||In the first part of the talk and motivated by an application in oil-field exploration, we consider decision problems in which the decision-maker can influence the time of information discovery. We formulate such problems as mixed-binary multi-stage stochastic programs with decision-dependent non-anticipativity constraints and propose a tractable solution scheme to compute near-optimal exploration strategies. We assess our approach on a problem of infrastructure and production planning in offshore oil fields from the literature.|
|+||In the second part of the talk and motivated by the recent explosion of data availability, we propose a data-driven paradigm for dynamic learning that unifies optimization and estimation. Our framework naturally captures the critical exploration-exploitation trade-off of the decision-maker, and we develop a tractable solution scheme to compute near-optimal policies. We assess our approach in the context of dynamic product pricing.|
|+||The first part of the talk is joint work with Daniel Kuhn and Berç Rustem, while the second part of the talk is joint work with Dimitris Bertsimas.|
|+||Phebe Vayanos is an Assistant Professor of Industrial and Systems Engineering at USC. Prior to joining USC, she was a Lecturer in the Operations Research and Statistics Group at MIT Sloan, and a postdoctoral research associate in the Operations Research Center at MIT. Her current research is focused on developing data-driven models and scalable solution approaches for real-world decision problems affected by uncertainty. In particular, she is motivated by applications in revenue management, energy, finance, education, and healthcare. She holds a PhD degree in Operations Research and an MEng degree in Electrical & Electronic Engineering, both from Imperial College London. She has extensive experience with the energy and investment banking industries, having worked at JPMorgan and BNP Paribas and having consulted for BP.|