I am currently working on two different problems, both of which require intensive computations, with additional GPU support being a valuable resource. The first problem involves communication efficient federated reinforcement learning (RL) and the second problem is on the study of offline RL. To support our study, we need to conduct extensive simulation studies for both problems. While the computations are both CPU and gpu-intensive. We will utilize neural networks in our models, which may benefit from additional GPU resources. Additionally, we will need to leverage multiple CPU cores in parallel.