The repository is for federated learning and transfer learning projects inside Huawei Noah's Ark Lab. Currently it includes:
- FairFL: a fair federated learning benchmarking framework that implements existing fair FL algorithms in the paper "Proportional Fairness in Federated Learning" by Guojun Zhang, Saber Malekmohammadi, Xi Chen and Yaoliang Yu. Comments are welcome!
- Hessian Alignment: domain generalization algorithms based on Hessian alignment (ICCV 2023). The related paper is "Understanding Hessian Alignment for Domain Generalization" by Sobhan Hemati*, Guojun Zhang*, Amir Estiri and Xi Chen. We have just released the code.
- Federated Domain Generalization: Incorporating domain generalization in the federated learning setting, based on the paper "Mitigating Data Heterogeneity in Federated Learning with Data Augmentation" by Artur Back de Luca*, Guojun Zhang*, Xi Chen, Yaoliang Yu.
- Layer Normalization: a federated learning benchmark that compares relevant FL algorithms for label shift problems, studied in the paper "Understanding the Role of Layer Normalization in Label-Skewed Federated Learning" by Guojun Zhang, Mahdi Beitollahi, Alex Bie, Xi Chen.
If you want to include your algorithms in one of our benchmarks, please make a pull request.
This open source project is not an official Huawei product, Huawei is not expected to provide support for this project.
