Katie Lu
Papers
2
Total Citations
19
H-Index
2
About
Katie Lu is a robotics researcher whose work focuses on the intersection of inverse reinforcement learning and active perception, with a particular emphasis on learning from human demonstrations. Her primary research areas include robotic ultrasound scanning, informative path-planning, and probabilistic temporal ranking for reward learning. Lu’s major contribution lies in developing novel frameworks that allow robots to infer task objectives from exploratory demonstrations, rather than requiring explicit cost functions or goal states. Her 2023 paper, "Learning rewards from exploratory demonstrations using probabilistic temporal ranking," which has garnered 15 citations, introduces a method for learning rewards in visual-servoing and active viewpoint selection tasks. This work builds on her earlier 2020 paper on robotic ultrasound scanning, where she demonstrated how robots can learn to adaptively search for satisfactory views by observing expert demonstrations. Lu’s research is particularly impactful for medical robotics, where precise, adaptive movements are critical. Her work has been recognized for its potential to enable robots to perform complex discovery tasks with minimal human input, making her a rising figure in the field of learning from demonstration.
Research Focus
Key Achievements
Top Papers
- 1
- 2