Kevin Jamieson
Papers
3
Total Citations
66
H-Index
2
About
Kevin Jamieson is a leading researcher at the intersection of robotics, machine learning, and human-robot interaction, with a core focus on how robots can learn efficiently from human demonstrations. His major contributions center on developing and comparing sampling strategies for robot deep learning, particularly in real-world, high-stakes environments like warehouse automation. In his seminal 2017 work, "Comparing human-centric and robot-centric sampling for robot deep learning from demonstrations," Jamieson systematically contrasted standard supervised learning from human teachers with robot-initiated queries, revealing critical trade-offs in learning efficiency and human effort—a study that has garnered 59 citations and shaped subsequent research in interactive learning. More recently, his 2023 paper on "Demonstrating Large-Scale Package Manipulation via Learned Metrics of Pick Success" tackles the practical challenge of automating logistics, introducing learned metrics that enable robots to reliably handle diverse packages at scale. This work directly addresses industry needs for resilient, cost-effective automation. Jamieson’s research is distinguished by its rigorous experimental methodology and its direct translation of theoretical insights into deployable robotic systems, making him a key figure in advancing data-efficient, human-aware robot learning.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3