Our paper FedQHD: Closed-Form Function-Space Federated Reinforcement Learning is now available on arXiv.