Federated Combinatorial Multi-Agent Multi-Armed Bandits

Authors: Fares Fourati, Mohamed-Slim Alouini, Vaneet Aggarwal

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

Reproducibility Variable Result LLM Response
Research Type Experimental We empirically validate our approach to a stochastic data summarization problem, illustrating the effectiveness of the proposed framework, even in single-agent scenarios.
Researcher Affiliation Academia 1Computer, Electrical and Mathematical Science and Engineering Division, King Abdullah University of Science and Technology (KAUST), Thuwal, KSA. 2School of Industrial Engineering, Purdue University, West Lafayette, IN 47907, USA. Correspondence to: Fares Fourati <fares.fourati@kaust.edu.sa>.
Pseudocode Yes We present our proposed C-MA-MAB Framework; see Algorithm 1.
Open Source Code No The paper does not provide any statement or link for open-source code for the methodology.
Open Datasets Yes We run experiments on FMNIST (Xiao et al., 2017) and CIFAR10 (Krizhevsky et al., 2009), present the latter in the main paper, and relegate more details and results to Appendix F.
Dataset Splits No The paper does not provide specific train/validation/test dataset splits. It mentions running experiments on FMNIST and CIFAR10, but no details on how these datasets were partitioned for training, validation, or testing.
Hardware Specification No The paper does not provide any specific hardware details used for running its experiments.
Software Dependencies No The paper does not provide specific software dependency details with version numbers.
Experiment Setup Yes We set a cardinality constraint of k = 5. Our goal is to summarize information from fifteen images, and instead of comparing it to all the images, we only consider a random batch R of 3 images. We run the experiments 100 times. We test for several time horizons in {125, 250, 500, 1000, 2000, 4000, 8000, 12000, 16000, 20000}.