Follow
Federated Q-Learning with Reference-Advantage Decomposition: Almost Optimal Regret and Logarithmic Communication Cost https://arxiv.org/abs/2405.18795 #stat.ML #cs.LG
Federated Q-Learning with Reference-Advantage Decomposition: Almost Optimal Regret and Logarithmic Communication Cost https://arxiv.org/abs/2405.18795 #stat.ML #cs.LG