Trustworthy Federated Learning Research

This repository contains research works and projects on trustworthy federated learning. It includes:

1. Datasets. Preprocessing codes of datasets we used and developed for federated learning research.
2. Publications. Implementation codes of our publications.
3. 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.
• FTL-FM. Grounding foundation models via federated transfer learning.
• 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 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

Title	Code	Description
Grounding Foundation Models through Federated Transfer Learning: A General Framework		Preprint
Optimizing Privacy, Utility and Efficiency in Constrained Multi-Objective Federated Learning		Preprint
Probably Approximately Correct 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
FedIPR: Ownership Verification for Federated Deep Neural Network Models	code	IEEE Transactions on Pattern Analysis and Machine Intelligence 2022
SecureBoost: A Lossless Federated Learning Framework	code	IEEE intelligent Systems 2021, widely-used federated tree-boosting algorithm
A Secure Federated Transfer Learning Framework	code	IEEE intelligent Systems 2020, the first federated transfer learning paper
FedVision: An Online Visual Object Detection Platform Powered by Federated Learning	code	AAAI 2020, Innovative Application of Artificial Intelligence Award from AAAI in 2020
Federated machine learning: Concept and applications		ACM TIST 2019, the 3rd most cited federated learning paper

FTL-FM

Title	Code	Description
Grounding Foundation Models through Federated Transfer Learning: A General Framework		Preprint
FATE-LLM: A Industrial Grade Federated Learning Framework for Large Language Models		LLM@IJCAI'23

Security and Privacy

Title	Code	Description
Achieving Provable Byzantine Fault-Tolerance in a Semi-honest Federated Learning Setting		PAKDD 2023
FedPass: Privacy-Preserving Vertical Federated Deep Learning with Adaptive Obfuscation		IJCAI 2023
A Framework for Evaluating Privacy-Utility Trade-off in Vertical Federated Learning		preprint
FedCG: Leverage Conditional GAN for Protecting Privacy and Maintaining Competitive Performance in Federated Learning	code	IJCAI 2022
Defending Batch-Level Label Inference and Replacement Attacks in Vertical Federated Learning	code	IEEE Transactions on Big Data
Secure Federated Matrix Factorization		IEEE Intelligent Systems 2020
Privacy-Preserving Deep Learning with SPDZ		The AAAI Workshop on PPAI
Batchcrypt: Efficient homomorphic encryption for cross-silo federated learning		USENIX 2020 ATC
Privacy Threats Against Federated Matrix Factorization		IJCAI 2020 FL workshop
Dynamic backdoor attacks against federated learning
Rethinking Privacy Preserving Deep Learning: How to Evaluate and Thwart Privacy Attacks		Springer Book 2020
Abnormal client behavior detection in federated learning		NIPS workshop 2019

Intellectual Property Protection

Title	Code	Description
FedTracker: Furnishing Ownership Verification and Traceability for Federated Learning Model		preprint
FedIPR: Ownership Verification for Federated Deep Neural Network Models		IEEE TPAMI, 2022
Protecting Intellectual Property of Generative Adversarial Networks from Ambiguity Attack		CVPR 2021
Rethinking Deep Neural Network Ownership Verification: Embedding Passports to Defeat Ambiguity Attack	code	NIPS 2019

Effectiveness

Title	Code	Description
FedHSSL: A Hybrid Self-Supervised Learning Framework for Vertical Federated Learning	code	IEEE Transactions on Big Data 2024
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		ACM TIST 2024
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: Concepts, Advances, and Challenges		TKDE 2024
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.

• 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

License

Apache 2.0 license.