Akshay L Chandra

prof_pic.png

Hi! I am a PhD Student at the Robot Learning Lab, supervised by Prof Dr. Abhinav Valada. At the moment, my research interests are at the intersection of reinforcement learning, world models and robot manipulation.

Recently, I completed M.Sc. Computer Science at the University of Freiburg. At the Robot Learning Lab, I worked almost two years as a Student Research Assistant on several robot manipulation problems, supervised by Dr. Iman Nematollahi and Dr. Tim Welschehold:

  • For Master’s Thesis. I fine-tuned diffusion-based policies inside world models to learn robot behaviours entirely offline [Paper, Code].
  • For Master’s Project, I fine-tuned geometry-aware (Riemannian manifold) motion models with model-free RL to stitch up to seven robot skills. [Report, Code].

Earlier (2018-2021), I was a full-time Research Assistant to Dr. Vineeth N Balasubramanian at the Machine Learning and Vision (MLV) research group in IIT Hyderabad, for nearly three years, mainly working on deep active learning for image data. During my time there, I also frequently collaborated with Dr. Wei Guo at the Field Phenomics Lab (University of Tokyo) and worked on several plant phenotyping solutions.