Tom Krumpolc
Tom Krumpolc CMU Chemical Engineering PhD Student

YinzOR 2023

YinzOR 2023

Hey there, Optimization enthusiasts! Buckle up for a whirlwind recap of the 6th annual YinzOR Student Conference, a two-day intellectual extravaganza hosted by the CMU INFORMS Student Chapter. Taking place at Carnegie Mellon University on August 25-26, 2023, the event was generously fueled by LocalSolver, the optimization software Maestros, whose support was nothing short of instrumental. The conference roared to life with opening remarks from Willem-Jan van Hoeve, the Carnegie Bosch Professor of Operations Research and Senior Associate Dean at Tepper CMU. His eloquent words set the stage for a cavalcade of enlightening talks from academia’s and industry’s finest. First up was Léa Blaise from LocalSolver, who wowed the crowd with her deep dive into LocalSolver’s modeling features, putting a spotlight on real-world scheduling problems. She was followed by Can Li from Purdue University, who unpacked the role of machine learning in global optimization, emphasizing its power to supercharge algorithms.

But that’s not all, folks! The first day also featured a poster competition that was a veritable intellectual showdown. With 13 participants from academic powerhouses like Georgia Tech, Temple University, University of California, Berkeley, McMaster University, Montréal Polytechnique, Northwestern University, and Carnegie Mellon University, the competition was a hotbed of innovation. Taking home the laurels were Wenbin Zhou from Carnegie Mellon University for his work on “Counterfactual Generative Models for Time-Varying Treatment,” Mo Liu from UC Berkeley for “Active Learning in the Predict-then-Optimize Framework: A Margin-Based Approach,” and Johnathan Vincente from Carnegie Mellon University for “Computing Equilibrium Automotive Technology Decisions Under Regularization,” with Hatice Gökçen Güner from Carnegie Mellon University winning the Fan Favorite award for “Machine Learning Driven Shape Optimization of Metamaterials with Adaptable Force-Displacement Characteristics for Soft Robots.” The esteemed panel of judges for this academic face-off included Cristiana Lara, Alan Scheller-Wolf, and Gérard Cornuéjols.

Following this intellectual feast, Cristiana Lara from Amazon took us on a journey through the labyrinthine world of inventory placement, offering a scalable method for tackling both head and tail demand. Rounding off the day was Michael Hamilton, who presented “Project 412: Bridging Students to Communities,” a compelling exploration of the challenges faced by Black-owned businesses in Pittsburgh and the untapped potential of university students as a market.

As the curtain fell on Day 1, attendees were treated to a gastronomic delight—an Italian dinner from PIADA, featuring a smorgasbord of pasta, salads, and beverages.

Day 2 kicked off with a hearty breakfast, setting the stage for a series of talks. Hansheng Jiang discussed “Intertemporal Pricing in the Presence of Consumer Behaviors,” spotlighting the sub-optimal nature of constant pricing and how considering consumer heterogeneity can improve retailer revenue. Holly Wiberg presented on “Mixed-Integer Optimization with Constraint Learning,” introducing a framework for optimization with learned constraints that is particularly applicable in healthcare.

Then came the Flash Talk Competition, a rapid-fire round of intellectual sparring. With 9 participants from schools like MIT, the University of Pittsburgh, Northwestern University, Cornell University, the University of California, Berkeley, Carnegie Mellon University, and Columbia University, the competition was fierce. Emerging victorious were Vikas Deep from Northwestern University for “Optimal Adaptive Experimental Design of Inference of Average Treatment Effect,” Miaolan Xie from Cornell University for “High Probability Complexity Bounds for Adaptive Optimization Methods with Stochastic Oracles,” and Amine Bennouna from MIT for “Holistic Robust Machine Learning and Data-Driven Decision-Making,” with Mo Liu from UC Berkeley snagging the Fan Favorite award for “Pricing under the Generalized Markov Chain Choice Model: Learning through Large-scale Click Behaviors.” The judges for this round were Michael Hamilton, Sridhar R. Tayur, and R. Ravi.

Post-competition, Emily Diana talked about “Balanced Filtering via Non-Disclosive Proxies,” focusing on the problem of non-disclosively collecting a sample of data that is balanced with respect to sensitive groups when group membership is unavailable or prohibited from use at collection time. Finally, Evelyn Gong, an incoming professor at Carnegie Mellon University, concluded with “Provably Optimal Reinforcement Learning for Inventory Problems with Unknown Cyclic Demands,” inventing reinforcement learning methods with provable optimality guarantees for inventory management problems.

The conference reached its crescendo with heartfelt closing remarks from Fatma Kilinc-Karzan, who perfectly encapsulated the spirit of YinzOR 2023, setting the stage for future research and innovation in Operations Research and related disciplines.

And there you have it—a comprehensive, yet lively, overview of YinzOR 2023. Until next year, keep those algorithms humming and those innovative ideas flowing!

YiznOR 2023 team of organizers: Nilsu Uzunlar (Co-Chair), Tian Wang (Co-Chair), Tom Krumpolc (Webmaster Co-Chair), Alex Lim (Webmaster Co-Chair), Lin An (Speaker Co-Chair), Mik Zlatin (Speaker Co-Chair), H. Satyam Verma (Marketing Chair), Siyue Liu (Flash Talk and Poster Session Co-Chair), Vrishabh Patil (Flash Talk and Poster Session Co-Chair), Macarena Navarro (Logistics Co-Chair), Sebastian Vasquez (Logistics Co-Chair), Anthony Karahalios (Sponsor Chair)