Research Interest

My journey started with pre-programmed robots. Soon, I began to wonder: what if these robots could offer more? This piqued my interest in the field of AI Robotics. While the practical benefits of intelligent robots are evident, I am most intrigued by what robotics can bring to the table to help us understand about general intelligence and ourselves!

The recent advancements in AI, largely driven by increased computational resources and time, come with inherent limitations. While it enables task-specific robots to excel in controlled environments, it doesn't equip them with the ability for meta-reasoning, which involves optimal allocation of computational resources, and meta-learning, where they can quickly adapt to new tasks using limited data. Such drawbacks impede their ability to safely and reliably operate in a human-unstructured environment and collaborate with us. Motivated by these challenges, my primary research interest lies in:

  • Robot Learning
  • Deep Reinforcement Learning
  • Human-robot Interaction

    • Looking ahead:
  • What approach will bootstrap the robot-learning process to be sample-efficient and application-agnostic?
    •