| Title | : | Robust Optimization using Machine Learning for Uncertainty Sets |
| Speaker | : | Theja Tulabandhula (Xerox Research Bangalore, India) |
| Details | : | Mon, 1 Dec, 2014 4:00 PM @ BSB 361 |
| Abstract: | : | Our goal is to build robust optimization problems that make decisions about the future, and where complex data from the past are used to model uncertainty. In robust optimization (RO) generally, the goal is to create a policy for decision-making that is robust to our uncertainty about the future. In particular, we want our policy to best handle the the worst possible situation that could arise, out of an textit{uncertainty set} of possible situations. Classically, the uncertainty set is simply chosen by the user, or it might be estimated in overly simplistic ways with strong assumptions; whereas in this work, we learn the uncertainty set from complex data from the past. The past data are drawn randomly from an (unknown) possibly complicated high-dimensional distribution. We propose a new uncertainty set design and show how tools from statistical learning theory can be employed to provide probabilistic guarantees on the robustness of the policy. Bio: Theja Tulabandhula has PhD in machine learning from CSAIL, MIT and B.Tech and M.Tech degrees in electrical engineering from IIT Kharagpur. He is currently a research scientist with Xerox Research Centre India. |
