ICLR 2024 accepted papers on Federated learning and privacy

Personalized and heterogeneous FL · 24 papers

• Fair and Efficient Contribution Valuation for Vertical Federated Learning (14 citations)
• Federated Recommendation with Additive Personalization (8 citations)
• Improving LoRA in Privacy-preserving Federated Learning (6 citations)
• Demystifying Local & Global Fairness Trade-offs in Federated Learning Using Partial Information Decomposition (4 citations)
• Benchmarking Algorithms for Federated Domain Generalization (3 citations)
• FedLoGe: Joint Local and Generic Federated Learning under Long-tailed Data (3 citations)
• VFLAIR: A Research Library and Benchmark for Vertical Federated Learning (3 citations)
• Fake It Till Make It: Federated Learning with Consensus-Oriented Generation (2 citations)
• FedImpro: Measuring and Improving Client Update in Federated Learning (2 citations)
• FedWon: Triumphing Multi-domain Federated Learning Without Normalization (2 citations)
• Internal Cross-layer Gradients for Extending Homogeneity to Heterogeneity in Federated Learning (2 citations)
• VertiBench: Advancing Feature Distribution Diversity in Vertical Federated Learning Benchmarks (2 citations)

and 12 more in the interactive map.

Differential privacy · 17 papers

• Privacy-Preserving In-Context Learning with Differentially Private Few-Shot Generation (8 citations)
• Privacy-Preserving In-Context Learning for Large Language Models (6 citations)
• One-shot Empirical Privacy Estimation for Federated Learning (5 citations)
• On Differentially Private Federated Linear Contextual Bandits (3 citations)
• Confidential-DPproof: Confidential Proof of Differentially Private Training (1 citation)
• Tractable MCMC for Private Learning with Pure and Gaussian Differential Privacy (1 citation)
• A Differentially Private Clustering Algorithm for Well-Clustered Graphs
• Correlated Noise Provably Beats Independent Noise for Differentially Private Learning
• Differentially Private SGD Without Clipping Bias: An Error-Feedback Approach
• Differentially Private Synthetic Data via Foundation Model APIs 1: Images
• DP-SGD Without Clipping: The Lipschitz Neural Network Way
• Efficiently Computing Similarities to Private Datasets

and 5 more in the interactive map.

Communication-efficient FL · 16 papers

• Adaptive Federated Learning with Auto-Tuned Clients (3 citations)
• Effective and Efficient Federated Tree Learning on Hybrid Data (2 citations)
• FedCompass: Efficient Cross-Silo Federated Learning on Heterogeneous Client Devices Using a Computing Power-Aware Scheduler (2 citations)
• Federated Wasserstein Distance (2 citations)
• Momentum Benefits Non-iid Federated Learning Simply and Provably (2 citations)
• FedHyper: A Universal and Robust Learning Rate Scheduler for Federated Learning with Hypergradient Descent (1 citation)
• Stochastic Controlled Averaging for Federated Learning with Communication Compression (1 citation)
• A Lightweight Method for Tackling Unknown Participation Statistics in Federated Averaging
• An improved analysis of per-sample and per-update clipping in federated learning
• Bayesian Coreset Optimization for Personalized Federated Learning
• Communication-Efficient Federated Non-Linear Bandit Optimization
• Federated Q-Learning: Linear Regret Speedup with Low Communication Cost

and 4 more in the interactive map.

Privacy attacks and secure learning · 5 papers

• DP-OPT: Make Large Language Model Your Privacy-Preserving Prompt Engineer (5 citations)
• Teach LLMs to Phish: Stealing Private Information from Language Models (3 citations)
• Be Careful What You Smooth For: Label Smoothing Can Be a Privacy Shield but Also a Catalyst for Model Inversion Attacks (2 citations)
• Hiding in Plain Sight: Disguising Data Stealing Attacks in Federated Learning
• FedInverse: Evaluating Privacy Leakage in Federated Learning