Thijs Wensveen
Delft University of Technology, RIKEN Center for Brain Science
Papers
2
Total Citations
10
H-Index
2
About
Thijs Wensveen’s research lies at the intersection of robotics, motor learning, and artificial intelligence, with a focus on enabling humanoid robots to learn more efficiently by mimicking biological systems. His major contribution is pioneering the concept of reusing previous experiences as simulation models to accelerate motor learning—a principle inspired by how animals adapt behavior through trial and error. In his most-cited work, “Trial and Error: Using Previous Experiences as Simulation Models in Humanoid Motor Learning” (2016, 6 citations), he demonstrated that robots can dramatically reduce the number of real-world samples needed to improve control policies by leveraging past data. His earlier study (2014, 4 citations) further validated this approach within a reinforcement learning framework, showing that humanoid robots could acquire nonlinear optimal policies with greater efficiency. Though his citation counts are modest, Wensveen’s work is notable for its conceptual elegance—bridging computational learning theory and biological cognition—and for addressing a fundamental bottleneck in real-world robotics: the high cost of physical trial and error. His research offers a promising pathway toward more adaptive, sample-efficient autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2Efficient reuse of previous experiences in humanoid motor learning4 citations · 2014