SIREN⁺: Robust Federated Learning with Proactive Alarming and Differential Privacy

This code accompanies the paper 'SIREN⁺: Robust Federated Learning with Proactive Alarming and Differential Privacy', which is accepted by IEEE Transactions on Dependable and Secure Computing (2024).

Please download Fashion-MNIST dataset to path like /home/data/ on the user's machine if you want to use it. CIFAR-10 dataset can be downloaded by the program automatically.

Recommended Dependencies: We have tested our system using the following environment configuration.

conda create -n siren-plus python=3.8.12
conda activate siren-plus
conda install tensorflow-gpu==2.4.1
pip install tensorflow-privacy==0.5.1
conda install keras==2.3.1 scikit-learn pandas

To run the code without any customized settings, please use:

python main.py

While if you want to run the code successfully with your customized parameters, you also need to set the following basic hyper-parameters:

Parameter	Function
--gar	Gradient Aggregation Rule
--eta	Learning Rate
--k	Number of agents
--C	Fraction of agents chosen per time step
--E	Number of epochs for each agent
--T	Total number of iterations
--B	Batch size at each agent
--mal_obj	Single or multiple targets
--mal_num	Number of targets
--mal_strat	Strategy to follow
--mal_boost	Boosting factor
--mal_E	Number of epochs for malicious agent
--ls	Ratio of benign to malicious steps in alt. min. attack
--rho	Weighting factor for distance constraint
--attack_type	attack type of malicious clients
--mia	use membership inference attack
--malicious_proportion	the proportion of malicious clients in the system
--non_iidness	the non-iidness of the data distribution on the clients

and SIREN⁺ exclusive parameters (if you want to use SIREN⁺):

Parameter	Function
--server_c	threshold $C_s$ used by the server
--client_c	threshold $C_c$ used by the client
--server_prohibit	threshold to trigger the penalty mechanism.
--forgive	the award value used by the award mechanism
--root_size	the size of the root test dataset
--dp	train local models with LDP on all the benign clients
--dp_mul	noise multiplier for LDP

For more parameters and details of the above parameters, please refer to \global_vars.py

If you want to use the same settings as us, here are some examples:

To run federated training with 10 agents and averaging based aggregation with Fashion-MNIST dataset, use

python main.py --dataset=fMNIST --k=10 --C=1.0 --E=5 --T=40 --B=64 --train --model_num=1 --gar=avg

While if you want to use CIFAR-10 dataset and ResNet-18, please set --dataset=CIFAR-10 and --model_num=9.

To run SIREN⁺ under label-flipping attack with Fashion-MNIST dataset using LDP, use

python main.py --dataset=fMNIST --k=10 --C=1.0 --E=5 --T=40 --B=64 --train --model_num=1 --mal --gar=siren --attack_type=label_flipping --dp --dp_mul=1.0

After running the code, please check /output directory for the results (please manually create the output directory before the execution of the codes).

To cite our paper, please use the following BibTex:

@article{guo2024siren+,
  title={SIREN+: Robust Federated Learning with Proactive Alarming and Differential Privacy},
  author={Guo, Hanxi and Wang, Hao and Song, Tao and Hua, Yang and Ma, Ruhui and Jin, Xiulang and Xue, Zhengui and Guan, Haibing},
  journal={IEEE Transactions on Dependable and Secure Computing (TDSC)},
  year={2024},
  publisher={IEEE}
}