Personalized Observation Normalization for Federated Reinforcement Learning in Simulation Environments with Heterogeneity
📰 ArXiv cs.AI
arXiv:2605.27385v1 Announce Type: cross Abstract: Federated reinforcement learning (FedRL) enables multiple agents to collaboratively train a global policy without sharing raw data, making it ideal for privacy-sensitive applications. However, FedRL faces challenges in heterogeneous environments where differing state-transition dynamics lead to non-identical input distributions and imbalanced parameter updates during aggregation. Therefore, this paper develops a personalized observation normaliza
Full Article
Title: Personalized Observation Normalization for Federated Reinforcement Learning in Simulation Environments with Heterogeneity
Abstract:
arXiv:2605.27385v1 Announce Type: cross Abstract: Federated reinforcement learning (FedRL) enables multiple agents to collaboratively train a global policy without sharing raw data, making it ideal for privacy-sensitive applications. However, FedRL faces challenges in heterogeneous environments where differing state-transition dynamics lead to non-identical input distributions and imbalanced parameter updates during aggregation. Therefore, this paper develops a personalized observation normaliza
Abstract:
arXiv:2605.27385v1 Announce Type: cross Abstract: Federated reinforcement learning (FedRL) enables multiple agents to collaboratively train a global policy without sharing raw data, making it ideal for privacy-sensitive applications. However, FedRL faces challenges in heterogeneous environments where differing state-transition dynamics lead to non-identical input distributions and imbalanced parameter updates during aggregation. Therefore, this paper develops a personalized observation normaliza
DeepCamp AI