Papers
38
Total Citations
2,165
H-Index
21
About
Raia Hadsell is a pioneering researcher at the intersection of deep learning, robotics, and artificial intelligence, whose work has fundamentally shaped how machines perceive, learn, and adapt to complex environments. Best known for her influential contributions to continual learning, her 2020 paper "Embracing Change: Continual Learning in Deep Neural Networks" (451 citations) has become a landmark reference for researchers grappling with how AI systems can learn incrementally without forgetting prior knowledge — a challenge central to building truly intelligent machines. Hadsell's robotics work spans autonomous navigation, robotic manipulation, and humanoid control. Her early research on self-supervised long-range terrain classification for off-road driving (316 citations) demonstrated the power of learning-based vision systems beyond stereo limitations. She has pushed the boundaries of reinforcement and imitation learning for visuomotor control, sim-to-real transfer, and even training bipedal robots to play soccer using deep RL. Her involvement in DeepMind's Gato project — a single generalist agent capable of multi-modal, multi-task performance — reflects her broader ambition to develop unified, adaptable AI systems. With over 1,500 cumulative citations across diverse domains, Hadsell stands as one of the field's most versatile and impactful voices.
Research Focus
Key Achievements
Top Papers
- 1Embracing Change: Continual Learning in Deep Neural Networks451 citations · 2020
- 2Learning long‐range vision for autonomous off‐road driving316 citations · 2009
- 3Reinforcement and Imitation Learning for Diverse Visuomotor Skills217 citations · 2018
- 4
- 5Reinforcement and Imitation Learning for Diverse Visuomotor Skills116 citations · 2018
- 6
- 7Sim-to-Real Robot Learning from Pixels with Progressive Nets109 citations · 2016
- 8
- 9A Generalist Agent66 citations · 2022
- 10One-Shot Reinforcement Learning for Robot Navigation with Interactive Replay53 citations · 2017