Theodore Tsitsimis

About

Theodore Tsitsimis is a researcher at the forefront of human-robot interaction (HRI), specializing in developing adaptive robotic systems that can respond to dynamic human behavior. His work centers on using reinforcement learning (RL) to enable robots to interpret and react to non-verbal cues—particularly human gaze and engagement—in real time. Tsitsimis’s major contribution is the integration of active exploration strategies within parameterized RL frameworks, allowing robots to autonomously adjust their behavior to maximize a human partner’s engagement. This is especially critical in sensitive applications like child-robot interaction, where individual preferences vary widely. His most cited paper (41 citations) demonstrates a robot’s ability to rapidly adapt to changes in human engagement during simulated social interaction, while another highly cited work (29 citations) formalizes the active exploration approach. Tsitsimis has also applied these methods to therapeutic and educational contexts for children with autism, where robots must tailor their social cues—such as eye contact—to each child’s comfort level. His research bridges machine learning and social robotics, offering a pathway toward more intuitive, responsive, and personalized robotic assistants.

Research Focus

Key Achievements

Top Papers

1. 1
   Robot Fast Adaptation to Changes in Human Engagement During Simulated Dynamic Social Interaction With Active Exploration in Parameterized Reinforcement Learning

   41 citations · 2018
2. 2
   Active Exploration and Parameterized Reinforcement Learning Applied to a Simulated Human-Robot Interaction Task

   29 citations · 2017
3. 3
   A Framework for Robot Learning During Child-Robot Interaction with Human Engagement as Reward Signal

   8 citations · 2018
4. 4
   Adaptive reinforcement learning with active state-specific exploration for engagement maximization during simulated child-robot interaction

   8 citations · 2018

Key Collaborators

Contact & Links