Idil Su Erdenlig

About

Idil Su Erdenlig’s research focuses on advancing reinforcement learning (RL) for complex, real-world systems, particularly in robotics and autonomous vehicles. Her major contribution lies in developing graph-based frameworks for hierarchical reinforcement learning (HRL), enabling agents to decompose complex tasks into manageable sub-problems. Her most-cited work, “Graph-Based Design of Hierarchical Reinforcement Learning Agents” (2019), introduces a novel approach to structuring multi-scale decision-making, addressing the critical challenge of scalability in RL. By leveraging graph theory, her method allows for more efficient learning and adaptation in dynamic environments, a key step toward deploying autonomous systems in unpredictable settings. While her citation count is still growing, her work has laid important groundwork for future HRL research, particularly in robotics and unmanned systems. Erdenlig’s research is notable for its emphasis on interpretable, modular architectures that bridge the gap between theoretical RL and practical deployment. Her contributions are especially relevant for students and researchers exploring hierarchical learning, multi-agent systems, or autonomous navigation, offering a principled way to design agents that can handle increasingly complex, real-world tasks.

Research Focus

Key Achievements

Top Papers

1. 1
   Graph-Based Design of Hierarchical Reinforcement Learning Agents

   2 citations · 2019

Key Collaborators

Contact & Links