README.md

FedMUD

The Panaceas for Improving Low-Rank Decomposition in Communication-Efficient Federated Learning [ICML 2025]

Environment Setup

Please execute the following command to install the necessary dependencies:

We recommend using uv to build your Python environment. If you're unfamiliar with uv， you can find an introduction and installation guide here.

uv sync
source .venv/bin/activate

Alternatively， you can also use command pip directly to install the necessary dependencies.

pip install -r requirements.txt

Data Partition

Run the following code to download and partition datasets:

cd dataloader/
python ./dataloader/datapartition.py

Run All Experiments

./run/ contains scripts to run the experiments, for example:

./run/12-feddmu.sh

Citation

@inproceedings{li2025fedmud,
  author       = {Shiwei Li and
                  Xiandi Luo and
                  Haozhao Wang and
                  Xing Tang and
                  Shijie Xu and
                  Weihong Luo and
                  Yuhua Li and
                  Xiuqiang He and
                  Ruixuan Li},
  title        = {The Panaceas for Improving Low-Rank Decomposition in Communication-Efficient Federated Learning.},
  booktitle    = {The Forty-Second International Conference on Machine Learning, {ICML 2025}, Vancouver, Canada, 13th-19th July, 2025},
  year         = {2025},
}