ML

I have been and am doing work at the Robot Intelligence through Perception Lab at TTIC.

My projects:

1. Active Advantage-Aligned Online Reinforcement Learning with Offline Data [link]
   • We propose a priority-based data sampling policy that improves on the uniform sampling of RLPD, by incorporating the onlineness of the the transitions and their estimated advantages.
2. Some more are in the oven!