Personalized modelsFairnessFederated learning (FL) is a framework for training machine learning models in a distributed and collaborative manner. During training, a set of participating clients process their data stored locally, sharing only the model updates obtained by minimizing a cost function over ...
In Federated Learning (FL), the clients learn a single global model (FedAvg) through a central aggregator. In this setting, the non-IID distribution of the data across clients restricts the global FL model from delivering good performance on the local data of each client. Personalized FL aims...
Ditto [21] measures the trade-off between robustness and fairness based on the design of a personalized model. Wang et al. [22] propose that gradient conflict is one aspect of affecting federated learning fairness, and use gradient surgery to reduce the conflict of each local gradient. The ...
@inproceedings{FedAS_CVPR24,author={Yang, Xiyuan and Huang, Wenke and Ye, Mang},title={FedAS: Bridging Inconsistency in Personalized Fedearated Learning},booktitle={CVPR},year={2024}}@article{FCCLPlus_TPAMI23,title={Generalizable Heterogeneous Federated Cross-Correlation and Instance Similarity ...
& Ding, B. pFL-Bench: a comprehensive benchmark for personalized federated learning. In 36th Conf. Neural Information Processing Systems Datasets and Benchmarks Track (2022). Chai, J. & Wang, X. Self-supervised fair representation learning without demographics. In Adv. Neural Information ...
Recent years have seen the rapid development of fairness-aware machine learning in mitigating unfairness or discrimination in decision-making in a wide range of applications. However, much less attention has been paid to the fairness-aware multi-objective optimization, which is indeed commonly seen in...
TitleAuthorsPublished inYearFilesNotesSupplementaries Extended PrivSR Towards privacy preserving social recommendation under personalized privacy settings Xuying Meng, Suhang Wang, Kai Shu, Jundong Li, Bo Chen, Huan Liu, Yujun Zhang WWWJ 2019 📒 FedMF Secure Federated Matrix Factorization Di Chai, ...
Personalized Federated Learning with Theoretical Guarantees: A Model-Agnostic Meta-Learning Approach Federated learning (FL) is currently the most widely adopted framework for collaborative training of (deep) machine learning models under privacy constrain... A Fallah,A Mokhtari,AE Ozdaglar - Neural Info...
You need first to go to ./data/cifar100, follow the instructions in README.md to download and partition the dataset. Average performance of personalized models Run the following scripts, this will generate tensorboard logs that you can interact with to make plots or get the values. # run ...
We adopt Ditto as our personalized FL approach, which enables each client in FL to develop its own personalized model through multi-task learning. Our assessment is achieved through a series of evaluations, comparing the predictive performance (i.e., AUROC and AUPRC) and fairness gaps (i.e....