Federated-Learning

A Hierarchical Gradient Tracking Algorithm for Mitigating Subnet-Drift in Fog Learning Networks

• Introduces SD‑GT, a semi‑decentralized gradient‑tracking method for fog‑based federated learning. • Eliminates gradient‑diversity assumptions, enabling robust training across hig

EMS-FL: Federated Tuning of Mixture-of-Experts in Satellite-Terrestrial Networks via Expert-Driven Model Splitting

• Combines Mixture‑of‑Experts (MoE) with satellite‑terrestrial networks (STN) to overcome data scarcity and compute limits in federated learning. • Introduces EMS‑FL, an expert‑dri

Federated Learning-Assisted Optimization of Mobile Transmission with Digital Twins

• Federated learning framework enables mobile transmission scheduling while preserving device privacy. • Three energy‑constrained problems tackled: minimize transmission time, fixe

Federated Reasoning Distillation Framework with Model Learnability-Aware Data Allocation

• Addresses bidirectional model learnability gap in federated LLM-SLM reasoning collaboration. • Introduces LaDa framework with learnability-aware data filter for high-reward sampl

Replication Study: Federated Text-Driven Prompt Generation for Vision-Language Models

• FedTPG introduces dynamic text-driven prompt generation for vision-language models in federated settings. • Replication evaluated on six datasets, achieving 74.58% seen, 76.00% u

Ringleader ASGD: The First Asynchronous SGD with Optimal Time Complexity under Data Heterogeneity

• Introduces Ringleader ASGD, the first asynchronous SGD achieving optimal time complexity under data heterogeneity. • Eliminates unrealistic similarity assumptions across workers'

Privacy-Preserving Federated Learning - Future Collaboration and Continued Research

• Final NIST blog concludes US‑UK collaboration on privacy‑preserving federated learning. • Series tracks evolution from theoretical PET discussions to real‑world adoption. • Explo

Data Pipeline Challenges of Privacy-Preserving Federated Learning

• PPFL hides raw data from training org, preventing quality assessment and format validation. • Traditional preprocessing steps are often omitted in PPFL research, focusing solely