The Princeton Workshop on Optimization, Learning, and Control is a single-track workshop highlighting the latest research advances across these disciplines. Its main goal is to foster new interactions and lay the groundwork for new collaborations.
Optimization has historically been crucial for solving problems in machine learning and control, such as training classifiers or designing real-time control systems. Thanks to the recent explosion in availability of data, ideas from machine learning and control are leading to exciting new directions in optimization. For example, learning-based heuristics have accelerated algorithms in nonconvex and combinatorial optimization. In a similar way, robust control methods have helped analyze the convergence behavior of optimization algorithms and foster new algorithmic developments. In addition, decision-focused learning techniques have recently integrated prediction and optimization tasks across a wide range of applications including vehicle routing, power systems scheduling, and predictive control of autonomous systems.
The goal of this workshop is to bring together researchers working at the intersection of optimization, learning, and control to explore creative synergies and foster a community across these areas.