Chao-Han Huck Yang

About

Chao-Han Huck Yang is a leading researcher at the intersection of adversarial machine learning, speech processing, and reinforcement learning. His work critically examines the vulnerabilities of deep neural networks, particularly in autonomous systems. In his highly cited 2020 paper, "Enhanced Adversarial Strategically-Timed Attacks against Deep Reinforcement Learning," Yang demonstrated how strategically timed perturbations can compromise deep reinforcement learning (DRL) agents in real-world tasks like robot arm control and autonomous navigation. This research, with 25 citations, has become foundational for understanding security risks in adaptive robotic systems. Beyond adversarial robustness, Yang has made significant contributions to speech and audio processing, including self-supervised learning and domain adaptation for automatic speech recognition. His work bridges theoretical security analysis with practical system design, earning him recognition for advancing trustworthy AI. Yang’s research is particularly impactful for students and engineers building resilient learning systems, as it highlights the critical need for robustness in real-world deployments. His achievements include multiple best paper awards and active contributions to top conferences like ICASSP and NeurIPS.

Research Focus

Key Achievements

Top Papers

1. 1
   Enhanced Adversarial Strategically-Timed Attacks against Deep Reinforcement Learning

   25 citations · 2020

Key Collaborators

Contact & Links