Mark Deisenroth
Papers
1
Total Citations
20
H-Index
1
About
Mark Deisenroth is a leading researcher in machine learning and robotics, best known for pioneering probabilistic approaches to robot control and reinforcement learning. His major contributions center on developing data-efficient algorithms that enable robots to learn complex tasks—such as grasping and manipulation—from limited, noisy interactions with uncertain environments. His highly cited work, including the seminal paper "Policy search for learning robot control using sparse data" (2014, 20 citations), introduced principled Bayesian methods that treat learning as a probabilistic inference problem, dramatically reducing the data required for effective policy learning. Deisenroth’s impact is evident in his foundational role in advancing model-based reinforcement learning, where his Gaussian process-based approaches have become standard tools for handling sparse, real-world data. Beyond his research, he has co-authored influential textbooks and served as a research director at DeepMind, shaping the next generation of AI and robotics researchers. His work continues to inspire students and practitioners seeking robust, sample-efficient solutions for autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1Policy search for learning robot control using sparse data20 citations · 2014