Trustworthy Federated Learning Research
This repository contains research works and projects on trustworthy federated learning. It includes:
- Datasets. Preprocessing codes of datasets we used and developed for federated learning research.
- Publications. Implementation codes of our publications.
- Projects. Other projects in federated learning.
Federated Learning Portal
This Federated Learning Portal keeps track of books, workshops, conference special tracks, journal special issues, standardization effort and other notable events related to the field of Federated Learning (FL).
Datasets
Dataset | Description |
---|---|
Street Dataset | A real-world object detection dataset that annotates images captured by a set of street cameras based on object present in them, including 7 object categories. |
Fed_ModelNet40 | It consists of images taken from various views of 3D models, and can be used for vertical federated learning research. |
NUS-WIDE | To simulate a vertical federated learning setting, the image features of samples is put on one party and the textual tags on another party. |
CheXpert | CheXpert is a large dataset of chest X-rays and can be used for vertical federated learning research. |
Publications
Our publications are categorized as below:
- Highlight. Papers that have high impact or we recommend to read.
- Security and Privacy. Security and privacy attacks and defenses.
- Intellectual Property Protection. Intellectual property protection and ownership verification (on model or data).
- Effectiveness. Various algorithms that aim to improve the effectiveness of FL.
- Efficiency. Communication and computation efficiency.
- Incentive. Incentive Mechanism.
- Theory. Theoretical work of federated learning.
- Application. Federated learning in real-world applications.
- Dataset. Datasets for federated learning research.
- Survey. Survey on various topics of federated learning.
High Citation Papers
Title | Code | Description | Semantic Scholar Citation | Google Scholar Citation (by 01/10/2023) |
---|---|---|---|---|
Federated machine learning: Concept and applications | ACM TIST 2019, the 3rd most cited federated learning paper | 2995 | ||
Advances and Open Problems in Federated Learning | Foundations and Trends in Machine Learning 2021 | 2711 | ||
SecureBoost: A Lossless Federated Learning Framework | code | IEEE intelligent Systems 2021, widely-used federated tree-boosting algorithm, best paper award | 333 | |
A Secure Federated Transfer Learning Framework | code | IEEE intelligent Systems 2020, the first federated transfer learning paper | 338 | |
FedVision: An Online Visual Object Detection Platform Powered by Federated Learning | code | AAAI 2020, Innovative Application of Artificial Intelligence Award from AAAI in 2020 | 144 | |
Batchcrypt: Efficient homomorphic encryption for cross-silo federated learning | code | 2020 USENIX ATC 2020 | 261 | |
A Fairness-aware Incentive Scheme for Federated Learning | AIES 2020 | 117 | ||
Rethinking Deep Neural Network Ownership Verification: Embedding Passports to Defeat Ambiguity Attack | code | NIPS 2019 | 118 | |
Towards Personalized Federated Learning | IEEE Transactions on Neural Networks and Learning Systems 2022 | 115 |
Highlight Paper
Security and Privacy
Intellectual Property Protection
Effectiveness
Title | Code | Description |
---|---|---|
FedHSSL: A Hybrid Self-Supervised Learning Framework for Vertical Federated Learning | code | Preprint |
FedPD: A Federated Learning Framework with Optimal Rates and Adaptivity to Non-IID Data | ICML FL workshop 2020 | |
A Secure Federated Transfer Learning Framework | code | IEEE intelligent Systems 2020 |
FedCVT: Semi-supervised Vertical Federated Learning with Cross-View Training | ACM TIST 2022 | |
Federated Transfer Reinforcement Learning for Autonomous Driving | code | Federated and Transfer Learning Book |
Privacy-preserving Heterogeneous Federated Transfer Learning | IEEE BigData 2019 | |
SecureBoost: A Lossless Federated Learning Framework | code | IEEE intelligent Systems 2021 |
Multi-Component Transfer Metric Learning for handling unrelated source domain samples | Knowledge-Based Systems | |
Cross-silo Federated Neural Architecture Search for Heterogeneous and Cooperative Systems | code | Federated and Transfer Learning Book |
Efficiency
Title | Code | Description |
---|---|---|
FLASHE: Additively Symmetric Homomorphic Encryption for Cross-Silo Federated Learning | Arxiv 2021 | |
Batchcrypt: Efficient homomorphic encryption for cross-silo federated learning | code | 2020 USENIX ATC 2020 |
FedBCD: A Communication-Efficient Collaborative Learning Framework for Distributed Features | code | IEEE Transactions on Signal Processing 2022 |
RPN: A Residual Pooling Network for Efficient Federated Learning | ECAI 2020 | |
Secure and Efficient Federated Transfer Learning | IEEE BigData 2019 |
Incentive
Title | Code | Description |
---|---|---|
Contribution-Aware Federated Learning for Smart Healthcare | IAAI 2022 | |
A Fairness-aware Incentive Scheme for Federated Learning | AIES 2020 | |
A Sustainable Incentive Scheme for Federated Learning | IEEE Intelligent Systems | |
A multi-player game for studying federated learning incentive schemes | IJCAI 2020 |
Theory
Title | Code | Description |
---|---|---|
Probably Approximately Correct Federated Learning | Preprint | |
A Game-theoretic Framework for Federated Learning | Preprint | |
Trading Off Privacy, Utility and Efficiency in Federated Learning | ACM TIST 2023 | |
No Free Lunch Theorem for Security and Utility in Federated Learning | ACM TIST 2022 |
Application
Title | Code | Description |
---|---|---|
Amalur: Data Integration Meets Machine Learning | ICDE 2023 (Vision paper) | |
StarFL: Hybrid Federated Learning Architecture for Smart Urban Computing | ACM TIST 2021 | |
Variation-Aware Federated Learning with Multi-Source Decentralized Medical Image Data | IEEE Journal of Biomedical and Health Informatics 2020 | |
Fedml: A research library and benchmark for federated machine learning | code | NeurIPS 2020 FL workshop |
Federated Transfer Learning for EEG Signal Classification | code | IEEE EMBC 2020 |
Multi-Agent Visualization for Explaining Federated Learning | IJCAI 2019 | |
HHHFL: Hierarchical Heterogeneous Horizontal Federated Learning for Electroencephalography | IJCAI FL workshop 2020 | |
FedVision: An Online Visual Object Detection Platform Powered by Federated Learning | code | AAAI 2020 |
Fair and Explainable Dynamic Engagement of Crowd Workers | IJCAI 2019 | |
Privacy-Preserving Technology to Help Millions of People: Federated Prediction Model for Stroke Prevention | IJCAI 2020 FL workshop |
Dataset
Title | Code | Description |
---|---|---|
Real-World Image Datasets for Federated Learning | code | NIPS FL workshop 2019 |
Survey
Title | Code | Description |
---|---|---|
Vertical Federated Learning | Preprint | |
A Survey on Heterogeneous Federated Learning | Preprint | |
Toward Responsible AI: An Overview of Federated Learning for User-centered Privacy-preserving Computing | ACM TIST 2022 | |
Towards Personalized Federated Learning | IEEE Transactions on Neural Networks and Learning Systems | |
Advances and Open Problems in Federated Learning | Foundations and Trends in Machine Learning 2021 | |
Threats to Federated Learning: A Survey | IJCAI FL workshop 2020 | |
Towards Utilizing Unlabeled Data in Federated Learning: A Survey and Prospective | ArXiv 2020 | |
Federated machine learning: Concept and applications | ACM TIST 2019 |
Projects
Currently, we are actively contributing to two projects, FedML (research-origented) and FATE (application-oriented).
FedML
FedML (Federated Machine Learning) is a research-oriented Federated Learning Library. It provides a plenty of out-of-the-box modules in federated learning, which greatly facilitates the development of new federated learning algorithms for researchers. We are co-contributor to this project and mainly maintain the part of vertical federated learning.
FATE
FATE (Federated AI Technology Enabler) is an industrial grade Federated Learning framework. It has already incorporated many of our proposed methods and algorithms to enhance its security and efficiency under various federated learning scenarios. Some of the implemented algorithms are listed below:
- SecureBoost: A Lossless Federated Learning Framework
- A Secure Federated Transfer Learning Framework
- A Communication Efficient Collaborative Learning Framework for Distributed Features
- Batchcrypt: Efficient homomorphic encryption for cross-silo federated learning