Event Category: Operations Research Seminar Series

4/20/23 | 4:15pm | E25-111         Nicole Immorlica Senior Principal Researcher Microsoft Research Abstract: A prevalent assumption in auction theory is that the auctioneer has full control over the market and that the allocation she dictates is final. In practice, however, agents might be able to resell acquired items in an aftermarket.…

4/13/23 | 4:15pm | E25-111         Yash Kanoria Associate Professor Columbia Abstract: Motivated by online matching markets and network revenue management (NRM) problems with many types (e.g., fulfillment optimization), we study dynamic spatial matching (DSM) in which supply and demand live in d dimensional space and need to be matched with each other…

3/23/23 | 4:15pm | E25-111         Hongseok Namkoong Associate Professor Columbia Abstract: Inferring causal relationships is critical to reliable decision-making. However, traditional modeling assumptions that allow adjusting prediction models to learn counterfactuals rarely hold in practice. Observed decisions depend on unrecorded confounding variables, user behavior shifts across space and time, and marginalized…

3/16/23 | 4:15pm | E25-111         Martin Wainwright Cecil H. Green Professor MIT Abstract: Stochastic approximation (SA) methods, dating back to the seminal work of Robbins-Monro (1951), are used to solve fixed point equations based on noisy observations. They are a computational workhorse in statistics, machine learning, and operations research. Classical results…

3/9/23 | 4:15pm | E25-111         Mike Pinedo Julius Schlesinger Professor of OM NYU Stern Abstract: Efficient scheduling of industrial systems typically has a major impact on productivity levels. In this seminar we focus on the applications of scheduling heuristics in two important industries, namely steelmaking and microelectronics. In steel production the…

3/2/23 | 4:15pm | E25-111         Amine Bennouna PhD Student (ORC Best Student Paper Award Winner) MIT Abstract: We study the design of data-driven decision-making and machine learning methods that enjoy a guaranteed out-of-sample performance. Our objective is to identify the best-performing method that meets a certain desired level of robustness. We…

2/23/23 | 4:15pm | E25-111         Katya Scheinberg Professor Cornell Abstract: Recently a variety of stochastic variants of adaptive methods have been developed and analyzed. These include stochastic step search, trust region and cubicly regularized Newton methods. Such methods adapt the step size parameter and use it to dictate the accuracy required…

2/16/23 | 4:15pm | E25-111         Greg Valiant Associate Professor Stanford Abstract: In many natural high-dimensional optimization settings, there are significant gaps between the amount of data or number of iterations required by algorithms whose memory usage scales linearly with the dimension versus more complex and memory-intensive algorithms. Do some problems inherently…

2/9/23 | 4:15pm | E25-111         Robert Kleinberg Professor Cornell Abstract: Consider a seller experimenting with posted prices to estimate the distribution of consumers’ willingness to pay, or a participant in an advertising auction varying their bid to learn the distribution of winning bids. Both of these parties face the problem of…

10/27/22 | 4:15pm | E51-145         Soroush Saghafian Associate Professor Harvard University Abstract: A main research goal in various applications is to use an observational data set and provide a new set of counterfactual guidelines that can yield causal improvements. When using available methods, researchers often have to rely on assumptions that…