Abstract:In the context of federated learning, smart terminal devices inherently exhibit various factors such as personalized user behaviors, regional differences, and heterogeneous hardware configurations due to their unique data capturing capabilities. Consequently, they inevitably present non-independent and identically distributed (Non-IID) data during the training process, making it challenging for traditional federated learning methods to achieve desired model performance and convergence effects when dealing with such complex data distributions. To address this challenge, researchers have explored data augmentation techniques, adaptive optimization algorithms, and improved model aggregation rules. However, these methods often fail to deliver satisfactory performance when confronted with Non-IID problems. In this paper, we propose a federated learning framework based on a greedy algorithm (FedGA). Unlike traditional approaches that rely on global parameter averaging aggregation, FedGA progressively searches for partially optimal models and aggregates them to obtain the global model. We first gather client data information and finely classify all clients, employing two strategies to optimize client data distribution: (1) for clients with imbalanced quantities, we utilize data sampling methods to balance client data; (2) for clients with imbalanced class distributions, we introduce a complementary federated meta-learning method, called FedComMeta, to improve the data distribution of client groups. Experimental results demonstrate that in scenarios involving Non-IID data, FedGA achieves significant improvements in model performance compared to existing methods such as FedAvg, FedProx, and Astraea, validating the effectiveness and superiority of our approach in handling Non-IID data distributions.

FedGR: Genetic Algorithm and Relay Strategy Based Federated Learning

FedDGP: Disentangling Global and Personal Models for Federated Learning

FedEmb: A Vertical and Hybrid Federated Learning Algorithm using Network And Feature Embedding Aggregation

FedGK: Communication-Efficient Federated Learning through Group-Guided Knowledge Distillation

AFed: Algorithmic Fair Federated Learning

FedGR: Federated Learning with Gravitation Regulation for Double Imbalance Distribution.

FedAgg: Adaptive Federated Learning with Aggregated Gradients

FedRDS: Federated Learning on Non-IID Data Via Regularization and Data Sharing

FedAA: A Reinforcement Learning Perspective on Adaptive Aggregation for Fair and Robust Federated Learning

Federated Learning with Domain Generalization

Lazy Aggregation for Heterogeneous Federated Learning

FedAux: an Efficient Framework for Hybrid Federated Learning

FedGA: A Greedy Approach to Enhance Federated Learning with Non-IID Data

Acceleration of Federated Learning with Alleviated Forgetting in Local Training

FedSG: A Personalized Subgraph Federated Learning Framework on Multiple Non-IID Graphs

Adaptive Federated Learning on Non-IID Data with Resource Constraint

A Fair Federated Learning Framework with Reinforcement Learning.

FedGroup: Efficient Clustered Federated Learning via Decomposed Data-Driven Measure

FedGrav: An Adaptive Federated Aggregation Algorithm for Multi-institutional Medical Image Segmentation

FedAGA: A federated learning framework for enhanced inter-client relationship learning

Preconditioned Federated Learning