2 datasets found

We study federated contextual linear bandits, where M agents cooperate with each other to solve a global contextual linear bandit problem with the help of a central server. We...

DeceFL is a decentralized federated learning algorithm that eliminates the need for a central server and maintains convergence property.