Motoaki Kawanabe
Advanced Telecommunications Research Institute International
Papers
6
Total Citations
518
H-Index
3
About
Motoaki Kawanabe is a leading researcher at the intersection of machine learning, brain-computer interfaces, and robotics. His foundational work addresses the critical challenge of **covariate shift**—the problem of data distribution changes in non-stationary environments—which is essential for deploying robust machine learning systems in the real world. His highly cited book, *Machine Learning in Non-Stationary Environments* (with over 470 combined citations), provides the theoretical framework and algorithms that enable models to adapt when training and test data differ, a breakthrough for fields from autonomous systems to healthcare. Beyond theory, Kawanabe has pioneered practical applications in **brain-controlled smart homes**, developing a waypoint-based framework that integrates noninvasive BMI with domotics and robotics to assist aging populations and individuals with disabilities. His recent work extends to embodied AI, including zero-shot question answering for robots and pedestrian density prediction for efficient mobile robot navigation. With a career spanning rigorous algorithmic foundations to human-centered robotics, Kawanabe’s research continues to shape how machines learn, adapt, and interact intelligently with dynamic, real-world environments.
Research Focus
Key Achievements
Top Papers
- 1
- 2Machine Learning in Non-Stationary Environments208 citations · 2012
- 3
- 4Map-based Modular Approach for Zero-shot Embodied Question Answering3 citations · 2024
- 5Pedestrian Density Prediction for Efficient Mobile Robot Exploration3 citations · 2019
- 6