Fair Federated Learning via the Proportional Veto Core

Authors: Bhaskar Ray Chaudhury, Aniket Murhekar, Zhuowen Yuan, Bo Li, Ruta Mehta, Ariel D. Procaccia

ICML 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Finally, we demonstrate that Rank-Core Fed outperforms baselines in terms of fairness on different datasets.
Researcher Affiliation Academia 1University of Illinois Urbana-Champaign 2University of Chicago 3Harvard University.
Pseudocode Yes Algorithm 1 Computes a representative set of P; Algorithm 2 Rank-Core-Fed: Finds a θ that belongs to the proportional veto core of P
Open Source Code No The paper does not provide any explicit statement or link to open-source code for the described methodology.
Open Datasets Yes We evaluate our algorithm Rank-Core-Fed on rotated MNIST (Le Cun et al., 2010) and CIFAR-10 (Krizhevsky et al., 2009) datasets.
Dataset Splits No The paper does not explicitly provide training/validation/test dataset splits, specific percentages, or sample counts.
Hardware Specification No The paper does not explicitly describe the hardware used to run its experiments.
Software Dependencies No The paper mentions the use of CNN and VGG11 models but does not provide specific version numbers for software dependencies or libraries.
Experiment Setup Yes For MNIST, we use a CNN, which has two 5 5 convolution layers followed by two fully connected layers with Re LU activation. For CIFAR-10, we evaluate with a more complex network, VGG11 (Simonyan & Zisserman, 2014). In all our experiments, we define agent utility as M Lce, where Lce refers to the average cross entropy loss on the agent s local test data. We set M to be 1.0 in our experiments. For all baselines and our algorithm, we set the number of iterations of the global model update to 50.