FederatedScope: A Flexible Federated Learning Platform for Heterogeneity
FederatedScope is a flexible and comprehensive federated learning platform proposed for tackling the heterogeneity in real-world federated learning applicati...
Publications by Year
2022
FederatedScope-GNN: Towards a Unified, Comprehensive and Efficient Package for Federated Graph Learning
FederatedScope-GNN is an easy-to-use python package for federated graph learning. We built it upon FederatedScope so that the requirements for expressing fed...
Federated Matrix Factorization with Privacy Guarantee
Matrix factorization algorithms for recommender systems under both horizontal and vertical federated settings. Published in VLDB 2022.
2020
Improving Utility and Security of the Shuffler-based Differential Privacy
For frequency queries, introducing a new algorithm that achieves a better privacy-utility tradeoff via shuffling and a novel protocol that provides better pr...
Collecting and Analyzing Data Jointly from Multiple Services under Local Differential Privacy
Techniques for collecting data from each service independently and analyzing the data from multiple services jointly, with privacy guarantees. Published in V...
2019
DPSAaS: Multi-Dimensional Data Sharing and Analytics as Services under Local Differential Privacy
We design and demonstrate a lightweight middleware called DPSAaS, which provides differentially private data-sharing-and-analytics functionality as cloud ser...
Answering Multi-Dimensional Analytical Queries under Local Differential Privacy
Algorithms for answering multi-dimensional analytical (MDA) queries approximately under local differential privacy. Published in SIGMOD 2019.