The field of machine learning is moving towards a greater emphasis on fairness and privacy, with a focus on developing algorithms and frameworks that can balance competing objectives such as fairness, utility, and privacy. Recent research has highlighted the importance of considering the trade-offs between these objectives, and has proposed new methods for achieving fairness and privacy in a variety of settings, including federated learning, bandit algorithms, and graph clustering. Notable papers in this area include those that propose new algorithms for fair federated learning, such as Accurate Target Privacy Preserving Federated Learning, and those that develop frameworks for evaluating and comparing the performance of different algorithms, such as A Framework for Fair Evaluation of Variance-Aware Bandit Algorithms. Other papers have explored the application of fairness and privacy techniques to specific domains, such as medical imaging and criminal justice.
Advances in Fairness and Privacy in Machine Learning
Sources
Accurate Target Privacy Preserving Federated Learning Balancing Fairness and Utility
A Framework for Fair Evaluation of Variance-Aware Bandit Algorithms
Group-Sensitive Offline Contextual Bandits
FairAD: Computationally Efficient Fair Graph Clustering via Algebraic Distance
DP-FedPGN: Finding Global Flat Minima for Differentially Private Federated Learning via Penalizing Gradient Norm
Benchmarking Federated Learning Frameworks for Medical Imaging Deployment: A Comparative Study of NVIDIA FLARE, Flower, and Owkin Substra
Toward Unifying Group Fairness Evaluation from a Sparsity Perspective
FedOnco-Bench: A Reproducible Benchmark for Privacy-Aware Federated Tumor Segmentation with Synthetic CT Data
Happiness as a Measure of Fairness
Towards Selection of Large Multimodal Models as Engines for Burned-in Protected Health Information Detection in Medical Images
Variance-Aware Feel-Good Thompson Sampling for Contextual Bandits
A Simple and Fast $(3+\varepsilon)$-approximation for Constrained Correlation Clustering
Improving the Performance of Radiology Report De-identification with Large-Scale Training and Benchmarking Against Cloud Vendor Methods
Fitting Reinforcement Learning Model to Behavioral Data under Bandits
Alternative Fairness and Accuracy Optimization in Criminal Justice