Optimization and Learning in Dynamic Systems

The field of dynamic systems is witnessing significant advancements with the integration of optimization and learning techniques. Researchers are exploring innovative approaches to address complex challenges in areas such as transportation, resource allocation, and humanitarian relief. Notably, reinforcement learning and deep learning algorithms are being applied to optimize dynamic tolling, refinery planning, and ride-pooling systems, yielding improvements in efficiency and performance. Furthermore, novel frameworks and algorithms are being developed to tackle issues like congestion, trust estimation, and generalized Nash equilibrium in multi-stage games. Some noteworthy papers include: Deep Reinforcement Learning for Day-to-day Dynamic Tolling in Tradable Credit Schemes, which achieves travel times and social welfare comparable to the Bayesian optimization benchmark. Iterative Recommendations based on Monte Carlo Sampling and Trust Estimation in Multi-Stage Vehicular Traffic Routing Games, which proposes a novel algorithm to compute the Bayesian Nash equilibrium and mitigate congestion.

Sources

Deep Reinforcement Learning for Day-to-day Dynamic Tolling in Tradable Credit Schemes

Reinforcement Learning-Driven Plant-Wide Refinery Planning Using Model Decomposition

Ride-pool Assignment Algorithms: Modern Implementation and Swapping Heuristics

Iterative Recommendations based on Monte Carlo Sampling and Trust Estimation in Multi-Stage Vehicular Traffic Routing Games

An Application of Membrane Computing to Humanitarian Relief via Generalized Nash Equilibrium

A Rollout-Based Algorithm and Reward Function for Efficient Resource Allocation in Business Processes

Built with on top of