Graph-Based Optimization and Control

The field of graph-based optimization and control is witnessing significant developments, with a focus on innovative numerical schemes and algorithmic frameworks. Researchers are exploring new methods to solve complex problems, such as Hamilton-Jacobi equations on Wasserstein spaces, discrete evacuation models, and multi-population Wardrop equilibria. These advancements have the potential to impact various applications, including traffic management, network design, and target interception. Noteworthy papers in this area include:

  • A study on finite difference schemes for Hamilton-Jacobi equations on Wasserstein spaces, which proposes novel numerical schemes and establishes their convergence and accuracy.
  • A paper on discrete evacuation in graphs with multiple exits, which provides an algorithmic framework for constructing valid evacuation strategies with constant competitive ratios.
  • A work on Hessian Riemannian flow for multi-population Wardrop equilibrium, which introduces a novel numerical method for optimizing flows on generalized graphs.
  • A research on traffic-oblivious multi-commodity flow network design, which presents a tight approximation algorithm for the Minimum Multi-Commodity Flow Subgraph problem.

Sources

Finite difference schemes for Hamilton--Jacobi equation on Wasserstein space on graphs

Discrete Evacuation in Graphs with Multiple Exits

Finite time max-consensus for simultaneous target interception in switching graph topologies

Hessian Riemannian Flow For Multi-Population Wardrop Equilibrium

Traffic-Oblivious Multi-Commodity Flow Network Design

Built with on top of