Paper-Conference

Federated Full-Parameter Tuning of Billion-Sized Language Models with Communication Cost under 18 Kilobytes
Federated Full-Parameter Tuning of Billion-Sized Language Models with Communication Cost under 18 Kilobytes

This work proposes a theory-informed method for federated full-parameter tuning of LLMs, which incurs <18KB communication cost per round for a 3B LLM, meanwhile delivering SOTA accuracy.

Jul 25, 2024
BlockDFL: A Blockchain-based Fully Decentralized Peer-to-Peer Federated Learning Framework
BlockDFL: A Blockchain-based Fully Decentralized Peer-to-Peer Federated Learning Framework

This work proposes a fully-decentralized P2P FL framework based on blockchain, which enables FL among a group of peer participants without mutual trust.

May 17, 2024
LARA: A Light and Anti-overfitting Retraining Approach for Unsupervised Time Series Anomaly Detection
LARA: A Light and Anti-overfitting Retraining Approach for Unsupervised Time Series Anomaly Detection

May 16, 2024
Learning Multi-Pattern Normalities in the Frequency Domain for Efficient Time Series Anomaly Detection
Learning Multi-Pattern Normalities in the Frequency Domain for Efficient Time Series Anomaly Detection

Mar 27, 2024
Resisting Backdoor Attacks in Federated Learning via Bidirectional Elections and Individual Perspective
Resisting Backdoor Attacks in Federated Learning via Bidirectional Elections and Individual Perspective

This work proposes a backdoor defense for FL from individual perspectives, which excludes infected model updates while maintaining main task accuracy.

Feb 28, 2024
FedAPEN: Personalized Cross-silo Federated Learning with Adaptability to Statistical Heterogeneity
FedAPEN: Personalized Cross-silo Federated Learning with Adaptability to Statistical Heterogeneity

This work proposes a personalized FL approach for cross-silo scenarios, which achieves adaptability of the statistical heterogeneity among client-side data.

Aug 7, 2023
TD-EUA: Task-Decomposable Edge User Allocation with QoE Optimization
TD-EUA: Task-Decomposable Edge User Allocation with QoE Optimization

Dec 9, 2020
BlockDFL: A Blockchain-based Fully Decentralized Peer-to-Peer Federated Learning Framework
BlockDFL: A Blockchain-based Fully Decentralized Peer-to-Peer Federated Learning Framework

Jul 8, 2019