The goal of this project is to use reinforcement learning to build pixel-based agents that successfully zero-shot transfer from a simulated environment to reality, which is...
