20 August 2026

Photo: UHH Knowledge Technology

On 20.08.2026, our colleague Mohammad Ali Zamani successfully defended his doctoral thesis “Reinforcement Learning for Human-Robot Interaction: Emotion-Based Feedback, Intermediate Rewards, and Faster Learning”. We congratulate him on his graduation and wish all the best in his next steps! Below is a brief overview of his thesis:

This dissertation investigates adaptive learning for human–robot interaction, focusing on how robots can detect human feedback, incorporate intermediate feedback into high-level planning, and acquire new skills efficiently. The research examines the early detection of warning signals through speech emotion recognition, studies how sparse intermediate rewards can improve reinforcement-learning-based action planning, and investigates adaptive task simplification combined with replay-memory management to accelerate continuous robot learning while mitigating catastrophic forgetting. Together, these contributions aim to support safer, more efficient, and more adaptive robotic systems in human environments.

We thank the committee and colleagues for their support and look forward to seeing Mohammad’s future contributions to Human-Robot Interaction!