Andrew Moore
Papers
10
Total Citations
3,180
H-Index
9
About
Andrew Moore is a pioneering figure in machine learning and robotics, best known for his foundational work in locally weighted learning, reinforcement learning, and memory-based control. His landmark 1997 paper, "Locally Weighted Learning," has amassed over 1,683 citations, establishing a core methodology for non-parametric regression that remains influential in robotics and AI. Moore’s dissertation on efficient memory-based learning for robot control (291 citations) formalized the SAB framework, enabling systems to construct world models from sensor data without prior knowledge—a critical advance for autonomous robotics. He also introduced the Parti-game algorithm (262 citations), a variable-resolution reinforcement learning method for high-dimensional state spaces, and the Racing algorithm (201 citations) for efficient model selection in lazy learning. His work on minimizing cross-validation error (228 citations) further advanced practical machine learning. More recently, Moore has shifted toward human-centered AI, exploring robotics to reduce hospital falls by enhancing patient-nurse interactions during toileting. With over 3,000 total citations, his research bridges theoretical rigor and real-world impact, inspiring generations of researchers in adaptive control and interactive AI systems.
Research Focus
Key Achievements
Top Papers
- 1Locally Weighted Learning1,683 citations · 1997
- 2Locally Weighted Learning302 citations · 1997
- 3Efficient memory-based learning for robot control291 citations · 2021
- 4
- 5Efficient Algorithms for Minimizing Cross Validation Error228 citations · 1994
- 6The Racing Algorithm: Model Selection for Lazy Learners201 citations · 1997
- 7
- 8The Racing Algorithm: Model Selection for Lazy Learners43 citations · 1997
- 9Q2: memory-based active learning for optimizing noisy continuous functions27 citations · 2002
- 10