NeurIPS 2024 accepted papers on Federated learning and privacy
114 of the 4,560 papers accepted at NeurIPS 2024 (2.5%) are on Federated learning and privacy, compared with 3.6% at NeurIPS 2023. They fall into 4 topics.
See NeurIPS 2024 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.
Topics and papers
Personalized and heterogeneous FL · 41 papers
- Federated Fine-tuning of Large Language Models under Heterogeneous Tasks and Client Resources (60 citations)
- FLoRA: Federated Fine-Tuning Large Language Models with Heterogeneous Low-Rank Adaptations (36 citations)
- Personalized Federated Learning via Feature Distribution Adaptation (31 citations)
- Classifier Clustering and Feature Alignment for Federated Learning under Distributed Concept Drift (21 citations)
- Bridging Gaps: Federated Multi-View Clustering in Heterogeneous Hybrid Views (17 citations)
- FedGMKD: An Efficient Prototype Federated Learning Framework through Knowledge Distillation and Discrepancy-Aware Aggregation (15 citations)
- Confusion-Resistant Federated Learning via Diffusion-Based Data Harmonization on Non-IID Data (14 citations)
- Dual-Personalizing Adapter for Federated Foundation Models (13 citations)
- FIARSE: Model-Heterogeneous Federated Learning via Importance-Aware Submodel Extraction (12 citations)
- Federated Model Heterogeneous Matryoshka Representation Learning (11 citations)
- FedLLM-Bench: Realistic Benchmarks for Federated Learning of Large Language Models (11 citations)
- Personalized Federated Learning with Mixture of Models for Adaptive Prediction and Model Fine-Tuning (11 citations)
and 29 more in the interactive map.
Differential privacy · 32 papers
- Prior-itizing Privacy: A Bayesian Approach to Setting the Privacy Budget in Differential Privacy (9 citations)
- Scalable DP-SGD: Shuffling vs. Poisson Subsampling (4 citations)
- Universal Exact Compression of Differentially Private Mechanisms (3 citations)
- Differentially Private Optimization with Sparse Gradients (2 citations)
- Differentially Private Reinforcement Learning with Self-Play (2 citations)
- Nearly Tight Black-Box Auditing of Differentially Private Machine Learning (2 citations)
- The Limits of Differential Privacy in Online Learning (2 citations)
- A Huber Loss Minimization Approach to Mean Estimation under User-level Differential Privacy (1 citation)
- Credit Attribution and Stable Compression (1 citation)
- Differential Privacy in Scalable General Kernel Learning via $K$-means Nystr{\"o}m Random Features (1 citation)
- Differentially Private Graph Diffusion with Applications in Personalized PageRanks (1 citation)
- Differentially Private Stochastic Gradient Descent with Fixed-Size Minibatches: Tighter RDP Guarantees with or without Replacement (1 citation)
and 20 more in the interactive map.
Communication-efficient FL · 23 papers
- Convergence Analysis of Split Federated Learning on Heterogeneous Data (15 citations)
- Heterogeneity-Guided Client Sampling: Towards Fast and Efficient Non-IID Federated Learning (15 citations)
- Hierarchical Federated Learning with Multi-Timescale Gradient Correction (11 citations)
- SpaFL: Communication-Efficient Federated Learning With Sparse Models And Low Computational Overhead (10 citations)
- Federated Learning under Periodic Client Participation and Heterogeneous Data: A New Communication-Efficient Algorithm and Analysis (8 citations)
- A-FedPD: Aligning Dual-Drift is All Federated Primal-Dual Learning Needs (4 citations)
- Revisiting Ensembling in One-Shot Federated Learning (4 citations)
- Communication-Efficient Federated Group Distributionally Robust Optimization (3 citations)
- Efficient Federated Learning against Heterogeneous and Non-stationary Client Unavailability (3 citations)
- Nonconvex Federated Learning on Compact Smooth Submanifolds With Heterogeneous Data (3 citations)
- Why Go Full? Elevating Federated Learning Through Partial Network Updates (3 citations)
- Local Superior Soups: A Catalyst for Model Merging in Cross-Silo Federated Learning (2 citations)
and 11 more in the interactive map.
Privacy attacks and secure learning · 18 papers
- RFLPA: A Robust Federated Learning Framework against Poisoning Attacks with Secure Aggregation (26 citations)
- Dual Defense: Enhancing Privacy and Mitigating Poisoning Attacks in Federated Learning (15 citations)
- DAGER: Exact Gradient Inversion for Large Language Models (10 citations)
- Attack-Aware Noise Calibration for Differential Privacy (9 citations)
- Open LLMs are Necessary for Current Private Adaptations and Outperform their Closed Alternatives (5 citations)
- SPEAR: Exact Gradient Inversion of Batches in Federated Learning (4 citations)
- DataStealing: Steal Data from Diffusion Models in Federated Learning with Multiple Trojans (3 citations)
- Ferrari: Federated Feature Unlearning via Optimizing Feature Sensitivity (3 citations)
- Parameter Disparities Dissection for Backdoor Defense in Heterogeneous Federated Learning (3 citations)
- Auditing Privacy Mechanisms via Label Inference Attacks (2 citations)
- Data Acquisition via Experimental Design for Data Markets (2 citations)
- FACT or Fiction: Can Truthful Mechanisms Eliminate Federated Free Riding? (2 citations)
and 6 more in the interactive map.