Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..
FedMut: Generalized Federated Learning via Stochastic Mutation
Authors: Ming Hu, Yue Cao, Anran Li, Zhiming Li, Chengwei Liu, Tianlin Li, Mingsong Chen, Yang Liu
AAAI 2024 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experimental results on well-known datasets demonstrate the effectiveness of our Fed Mut approach in various data heterogeneity scenarios. |
| Researcher Affiliation | Academia | 1School of Computer Science and Engineering, Nanyang Technological University, Singapore 2Mo E Engineering Research Center of SW/HW Co-Design Tech. and App., East China Normal University, China |
| Pseudocode | Yes | Algorithm 1: Implementation of Fed Mut |
| Open Source Code | No | The paper does not provide an explicit statement or link for the open-sourcing of its own code. |
| Open Datasets | Yes | We selected three well-known datasets to evaluate the effectiveness of our Fed Mut approach, i.e., CIFAR-10, CIFAR-100 (Krizhevsky 2009), and Shakespeare (Caldas et al. 2018), respectively, where CIFAR-10 and CIFAR-100 are image datasets and Shakespeare is a text dataset. |
| Dataset Splits | No | The paper describes training and testing, but it does not explicitly specify a validation dataset split or its size/percentage. |
| Hardware Specification | Yes | We conducted all the experiments on an Ubuntu workstation with an Intel i9 CPU, 64GB memory, and two NVIDIA RTX 4090 GPUs. |
| Software Dependencies | No | The paper mentions software components like 'SGD optimizer' and models like 'Res Net-18' and 'VGG16 (Torchvision Model 2022)', but does not specify versions for programming languages or libraries (e.g., Python, PyTorch). |
| Experiment Setup | Yes | For all the experiments, we set SGD optimizer with a learning rate of 0.01 and a momentum of 0.9. For each FL training round, we set the batch size to 50 and the number epoch of each local training to 5. |