Biography

Hello! I am Runyu Lu (陆润宇), a Ph.D. Candidate at the University of Chinese Academy of Sciences (UCAS), specializing in Computer Applied Technology at the School of Artificial Intelligence (Institute of Automation). My expected graduation date is June 2026.

My research is centered around Game Theory and Deep Reinforcement Learning (DRL), with a focus on developing robust and theoretically grounded algorithms for complex multi-agent systems. I am particularly interested in equilibrium learning theory and its application in real-world adversarial settings.

I hold a B.Eng. in Computer Science and Technology from the University of Chinese Academy of Sciences (2021).

Academic Highlights

Research Focus

My primary goal is to develop model-based, online, or offline learning algorithms with last-iterate convergence guarantees towards approximate or exact Nash equilibria, facilitating stable implementation in deep RL contexts. Key research themes include:

  • Equilibrium Finding Theory and Last-Iterate Convergence.
  • Multi-Agent Reinforcement Learning (MARL) in Dynamic Games.
  • Equilibrium Policy Generalization (EPG) for Cross-Domain Robustness.
  • Worst-Case Robust Real-Time Strategies (R2PS) under Partial Observability.

Education

  • Ph.D. Candidate (2021 – Expected 2026): Computer Applied Technology, UCAS.     * Advisor: Prof. Dongbin Zhao.     * Dissertation Topic: Equilibrium Finding Theory and Reinforcement Learning for Dynamic Games.
  • B.Eng. (2017 – 2021): Computer Science and Technology, UCAS.     * GPA: 3.92/4.00.

Awards

  • 2025: National Scholarship for Doctoral Students, University of Chinese Academy of Sciences.
  • 2025: IEEE CIS Graduate Student Research Grant, IEEE Computational Intelligence Society.
  • 2018: Silver Medal, ICPC Asia-East Continent Final, Association for Computing Machinery.
  • 2018: Silver Medal, ACM-ICPC Asia Regional Contest Beijing Site, Association for Computing Machinery (ACM).

Technical Skills

Category Expertise
Programming C++, Python, and LLMs
Core Expertise Reinforcement Learning, Game Theory, Algorithm Design, Multi-Agent Systems