Research lies at the intersection of machine learning and optimization, aiming to enable automated decision-making with AI for complex optimization problems.
Core research focuses on 'Learning to Optimize', leveraging advanced ML techniques to build deep models that assist or directly solve challenging optimization problems.
Primary PhD research centers on Neural Combinatorial Optimization (NCO), especially improving generalization of neural solvers, with applications to routing problems like TSP and VRP.
Recent work addresses practical combinatorial optimization (CO) challenges, including handling complex constraints, building robust and trustworthy CO approaches via human-AI collaboration, and advancing foundation models for combinatorial optimization (FM4CO).
Research Keywords: Vehicle Routing; Neural Combinatorial Optimization; Learning to Optimize