Ditto: Fair and Robust Federated Learning Through Personalization

2020/12/08 by Tian Li, Shengyuan Hu, Li, Tian +5 · 51 citations
Computer Science · #Cryptography and Data Security #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Privacy-Preserving Technologies in Data #Stochastic Gradient Optimization Techniques

paper · pdf · doi:10.48550/arxiv.2012.04221

openalex publication_date 2020/12/08 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Fairness and robustness are two important concerns for federated learning systems. In this work, we identify that robustness to data and model poisoning attacks and fairness, measured as the uniformity of performance across devices, are competing constraints in statistically heterogeneous networks. To address these constraints, we propose employing a simple, general framework for personalized federated learning, Ditto, that can inherently provide fairness and robustness benefits, and develop a scalable solver for it. Theoretically, we analyze the ability of Ditto to achieve fairness and robustness simultaneously on a class of linear problems. Empirically, across a suite of federated datasets, we show that Ditto not only achieves competitive performance relative to recent personalization methods, but also enables more accurate, robust, and fair models relative to state-of-the-art fair or robust baselines.

Citations

• Poisoning Attacks against Support Vector Machines
• EMNIST: an extension of MNIST to handwritten letters
• BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chain
• Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoning
• LEAF: A Benchmark for Federated Settings
• Improving Federated Learning Personalization via Model Agnostic Meta Learning
• Federated Meta-Learning with Fast Convergence and Efficient Communication
• Advances and Open Problems in Federated Learning
• Robust Aggregation for Federated Learning
• Personalized Federated Learning: A Meta-Learning Approach
• Three Approaches for Personalization with Applications to Federated Learning
• Personalized Federated Learning with Moreau Envelopes
• Federated Learning Meets Multi-objective Optimization
• Federated Reconstruction: Partially Local Federated Learning

Cited by

• FairGFL: Privacy-Preserving Fairness-Aware Federated Learning with Overlapping Subgraphs
• Evidential Trust-Aware Model Personalization in Decentralized Federated Learning for Wearable IoT
• DFedReweighting: A Unified Framework for Objective-Oriented Reweighting in Decentralized Federated Learning
• Federated style aware transformer aggregation of representations
• TOFA: Training-Free One-Shot Federated Adaptation for Vision-Language Models
• Personalized Federated Learning with Bidirectional Communication Compression via One-Bit Random Sketching
• Divide, Conquer and Unite: Hierarchical Style-Recalibrated Prototype Alignment for Federated Medical Segmentation
• FedeCouple: Fine-Grained Balancing of Global-Generalization and Local-Adaptability in Federated Learning
• EPFL-REMNet: Efficient Personalized Federated Digital Twin Towards 6G Heterogeneous Radio Environment
• Federated Quantum Kernel Learning for Anomaly Detection in Multivariate IoT Time-Series
• Communication-Constrained Private Decentralized Online Personalized Mean Estimation
• Iterative Foundation Model Fine-Tuning on Multiple Rewards
• Local Performance vs. Out-of-Distribution Generalization: An Empirical Analysis of Personalized Federated Learning in Heterogeneous Data Environments
• Sentinel: Dynamic Knowledge Distillation for Personalized Federated Intrusion Detection in Heterogeneous IoT Networks
• Federated Learning via Meta-Variational Dropout
• CO-PFL: Contribution-Oriented Personalized Federated Learning for Heterogeneous Networks
• Personalized Collaborative Learning with Affinity-Based Variance Reduction
• Parameter-Efficient and Personalized Federated Training of Generative Models at the Edge
• Incentivizing Time-Aware Fairness in Data Sharing
• FedL2T: Personalized Federated Learning with Two-Teacher Distillation for Seizure Prediction
• Locally Optimal Private Sampling: Beyond the Global Minimax
• MT-DAO: Multi-Timescale Distributed Adaptive Optimizers with Local Updates
• Beyond Static Knowledge Messengers: Towards Adaptive, Fair, and Scalable Federated Learning for Medical AI
• Personalized federated prototype learning in mixed heterogeneous data scenarios
• FeDABoost: Fairness Aware Federated Learning with Adaptive Boosting
• Beyond Aggregation: Guiding Clients in Heterogeneous Federated Learning
• Choice Outweighs Effort: Facilitating Complementary Knowledge Fusion in Federated Learning via Re-calibration and Merit-discrimination
• OmniFed: A Modular Framework for Configurable Federated Learning from Edge to HPC
• pFedSAM: Personalized Federated Learning of Segment Anything Model for Medical Image Segmentation
• FedHK-MVFC: Federated Heat Kernel Multi-View Clustering
• Who to Trust? Aggregating Client Knowledge in Logit-Based Federated Learning
• FedKLPR: Personalized Federated Learning for Person Re-Identification with Adaptive Pruning
• Bi-level Personalization for Federated Foundation Models: A Task-vector Aggregation Approach
• Reconciling Communication Compression and Byzantine-Robustness in Distributed Learning
• FedDAF: Federated Domain Adaptation Using Model Functional Distance
• Fed-REACT: Federated Representation Learning for Heterogeneous and Evolving Data
• AI-in-the-Loop: Privacy Preserving Real-Time Scam Detection and Conversational Scambaiting by Leveraging LLMs and Federated Learning
• Variational Gaussian Mixture Manifold Models for Client-Specific Federated Personalization
• Towards Instance-wise Personalized Federated Learning via Semi-Implicit Bayesian Prompt Tuning
• Federated Learning with Heterogeneous and Private Label Sets
• FedPrune: Towards Inclusive Federated Learning
• Deploying Models to Non-participating Clients in Federated Learning without Fine-tuning: A Hypernetwork-based Approach
• Breaking the Aggregation Bottleneck in Federated Recommendation: A Personalized Model Merging Approach
• Generalizable Federated Learning using Client Adaptive Focal Modulation
• APFL: Analytic Personalized Federated Learning via Dual-Stream Least Squares
• Flexible Personalized Split Federated Learning for On-Device Fine-Tuning of Foundation Models
• Don't Reach for the Stars: Rethinking Topology for Resilient Federated Learning
• FedHiP: Heterogeneity-Invariant Personalized Federated Learning Through Closed-Form Solutions
• FeDaL: Federated Dataset Learning for Time Series Foundation Models
• FedAPTA: Federated Multi-task Learning for Heterogeneous Devices with Adaptive Layer-wise Pruning and Task-aware Aggregation
• Boosting Generalization Performance in Model-Heterogeneous Federated Learning Using Variational Transposed Convolution

Related

• FedTilt: Towards Multi-Level Fairness-Preserving and Robust Federated Learning
• Fairness-Constrained Optimization Attack in Federated Learning
• Fairness and Accuracy in Federated Learning
• Federated Learning with Relative Fairness
• Certified Robustness in Federated Learning