Runze Liu
Papers
1
Total Citations
4
H-Index
1
About
Runze Liu is a rising researcher in the field of artificial intelligence security, with a primary focus on the robustness of deep reinforcement learning (DRL) systems. Their most notable contribution is the introduction of "RAT" (Reward-Attack Targeting), a framework for executing adversarial attacks that manipulate DRL agents into performing targeted, attacker-specified behaviors—a significant departure from traditional reward-based attacks. This work, published in 2025 and already garnering 4 citations, addresses a critical gap in evaluating agent security by demonstrating how adversaries can bypass conventional defenses to achieve precise behavioral outcomes. Liu’s research is pivotal for developing more resilient autonomous systems, particularly in high-stakes applications like robotics, gaming, and autonomous driving. By exposing vulnerabilities in DRL agents, Runze Liu is helping to shape the next generation of secure and trustworthy AI, making their work essential reading for students and researchers interested in adversarial machine learning and AI safety.
Research Focus
Key Achievements
Top Papers
- 1