Nikolas Hemion
Papers
10
Total Citations
101
H-Index
6
About
Nikolas Hemion is a robotics and artificial intelligence researcher whose work spans human-robot interaction, robot learning, and social robotics. His research is particularly focused on enabling robots to acquire new skills autonomously and through intuitive human guidance, as well as on making robots more socially engaging through expressive behavior. Among his most notable contributions is his work on generating emotional body language in humanoid robots using variational autoencoders, exploring how variation and complexity in robotic expression can sustain user engagement during long-term interaction (21 citations). His investigations into robot skill learning through naive user feedback—demonstrated with the Pepper robot—highlight his commitment to making personal robotics accessible to non-expert users (12 citations). He has also advanced goal babbling techniques for online learning in high-dimensional robotic systems, broadening how robots can develop generalizable sensorimotor skills from real-world experience. With contributions spanning hierarchical reinforcement learning as a framework for creative problem solving (23 citations) and developmental approaches to sensorimotor contingencies, Hemion's research bridges cognitive science and practical robotics. His accumulated body of work reflects a sustained effort to build robots that learn flexibly, interact naturally, and integrate meaningfully into human environments.
Research Focus
Key Achievements
Top Papers
- 1Hierarchical reinforcement learning as creative problem solving23 citations · 2016
- 2Generating robotic emotional body language with variational autoencoders21 citations · 2019
- 3
- 4
- 5
- 6
- 7
- 8Data-driven emotional body language generation for social robotics4 citations · 2022
- 9
- 10