Khimya Khetarpal
Papers
3
Total Citations
213
H-Index
3
About
Khimya Khetarpal is a leading researcher at the intersection of reinforcement learning (RL) and continual learning, dedicated to building AI systems that can learn and adapt over a lifetime. Her primary research areas include continual reinforcement learning, lifelong learning, and non-stationary environments. Khetarpal’s major contribution is her comprehensive review, "Towards Continual Reinforcement Learning: A Review and Perspectives," which has garnered over 200 citations. This seminal work systematically categorizes formulations and approaches for enabling RL agents to learn continuously without forgetting, establishing a foundational framework for the field. She argues persuasively that RL is a natural paradigm for studying continual learning due to its sequential decision-making nature. Khetarpal’s impact is evident in the widespread adoption of her taxonomies and perspectives by researchers tackling catastrophic forgetting and adaptation in dynamic settings. Earlier in her career, she explored mobile robot navigation using evolving neural controllers, demonstrating her versatility. Her work is essential reading for students and researchers aiming to build AI that learns as robustly and flexibly as biological systems.
Research Focus
Key Achievements
Top Papers
- 1Towards Continual Reinforcement Learning: A Review and Perspectives179 citations · 2022
- 2Towards Continual Reinforcement Learning: A Review and Perspectives28 citations · 2020
- 3