Collaborative Fairness

Collaborative fairness in federated learning aims to ensure that all participating clients contribute equitably to a shared model, receiving benefits proportional to their contribution while maintaining privacy. Current research focuses on developing methods to accurately assess individual client contributions, often leveraging game-theoretic concepts like Shapley values, and using these assessments to weight model updates or allocate resources dynamically. This work is crucial for improving the accuracy and robustness of federated learning models while incentivizing broader participation and addressing potential biases stemming from unequal data distributions across clients, with applications spanning diverse fields like healthcare and finance.