Learn-to-Optimize Frameworks for Networks: Models and Applications

• Network Optimization is a subclass of Combinatorial Optimization, a complex class of optimization problems, where the goal is to find the optimal solution from a finite candidate solution set. 

• Useful information can inform machine learning components to guide the search and expedite the exploration effort of optimization algorithms.

• Network features (spatial, temporal, and network measures) can be used to develop solution algorithms for networks.

• Integrate machine learning and optimization algorithms to develop scalable and customizable solution methods for large-scale networks.