Zheng Lin

About

Zheng Lin is a leading researcher at the intersection of artificial intelligence and robotics, with a primary focus on enhancing the robustness and safety of deep reinforcement learning (DRL) systems. Their most cited work, "Robust Deep Reinforcement Learning in Robotics via Adaptive Gradient-Masked Adversarial Attacks" (2025), tackles a critical bottleneck in real-world robotic deployment: the vulnerability of DRL policies to environmental perturbations. Lin’s key contribution lies in developing adaptive, gradient-masked adversarial attack methods that expose and fortify weaknesses in robotic control, moving beyond conventional white-box approaches that falter in dynamic settings. This research has garnered 3 citations in its early release, signaling growing influence in the field. By systematically stress-testing DRL agents, Lin’s work paves the way for more reliable autonomous systems in applications like manufacturing and service robotics. Their achievements underscore a commitment to bridging the gap between theoretical reinforcement learning and practical, resilient robotic intelligence—a vital step toward trustworthy AI in the physical world.

Research Focus

Key Achievements

Top Papers

1. 1
   Robust Deep Reinforcement Learning in Robotics via Adaptive Gradient-Masked Adversarial Attacks

   3 citations · 2025

Key Collaborators

Contact & Links