Investigating Local Differential Privacy and Generative Adversarial Network in Collecting Data

Ha, Trung; Dano, Tran Khanh

doi:10.1109/acomp50827.2020.00028

Cited by 4 publications

(2 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One of the clients who participated in the system was the adversary (Ha et al , 2019). The attacker performs two learnings: the first, they performed the main task, which took part in the federated training process to collaboratively train a model; the second, they performed the adversarial task, which made an inference attack based on the GAN model (Ha and Dang, 2020). In this study, the adversarial task looked closely at the ith participant as shown in Figure 11.…”

Section: Methodsmentioning

confidence: 99%

Inference attacks based on GAN in federated learning

Dang

2022

IJWIS

View full text Add to dashboard Cite

Purpose In the digital age, organizations want to build a more powerful machine learning model that can serve the increasing needs of people. However, enhancing privacy and data security is one of the challenges for machine learning models, especially in federated learning. Parties want to collaborate with each other to build a better model, but they do not want to reveal their own data. This study aims to introduce threats and defenses to privacy leaks in the collaborative learning model. Design/methodology/approach In the collaborative model, the attacker was the central server or a participant. In this study, the attacker is on the side of the participant, who is “honest but curious.” Attack experiments are on the participant’s side, who performs two tasks: one is to train the collaborative learning model; the second task is to build a generative adversarial networks (GANs) model, which will perform the attack to infer more information received from the central server. There are three typical types of attacks: white box, black box without auxiliary information and black box with auxiliary information. The experimental environment is set up by PyTorch on Google Colab platform running on graphics processing unit with labeled faces in the wild and Canadian Institute For Advanced Research-10 data sets. Findings The paper assumes that the privacy leakage attack resides on the participant’s side, and the information in the parameter server contains too much knowledge to train a collaborative machine learning model. This study compares the success level of inference attack from model parameters based on GAN models. There are three GAN models, which are used in this method: condition GAN, control GAN and Wasserstein generative adversarial networks (WGAN). Of these three models, the WGAN model has proven to obtain the highest stability. Originality/value The concern about privacy and security for machine learning models are more important, especially for collaborative learning. The paper has contributed experimentally to private attack on the participant side in the collaborative learning model.

show abstract

Section: Methodsmentioning

confidence: 99%

Inference attacks based on GAN in federated learning

Dang

2022

IJWIS

View full text Add to dashboard Cite

show abstract

“…Liu et al [ 60 ] proposed a GAN model for privacy protection, which achieves differential privacy by adding carefully designed noise to the clipping gradient in the process of model learning, uses the moment accountant strategy to improve the stability and compatibility of the model by controlling the loss of privacy, and generates high-quality synthetic data while retaining the required available data under a reasonable privacy budget. Ha and Dang [ 61 ] proposed a local differential privacy GAN model for noise data generation, which establishes a generative model by clipping the gradient in the model and adding Gaussian noise to the gradient to ensure the differential privacy. Chen et al [ 62 ] proposed gradient-sanitized WGAN, which allows the publication of sanitized sensitive data under strict privacy guarantee and can more accurately distort gradient information so as to train deeper models and generate more information samples.…”

Section: Differential Privacy Synthetic Data Generationmentioning

confidence: 99%

GDP vs. LDP: A Survey from the Perspective of Information-Theoretic Channel

Liu

Peng

Tian

et al. 2022

Entropy

View full text Add to dashboard Cite

The existing work has conducted in-depth research and analysis on global differential privacy (GDP) and local differential privacy (LDP) based on information theory. However, the data privacy preserving community does not systematically review and analyze GDP and LDP based on the information-theoretic channel model. To this end, we systematically reviewed GDP and LDP from the perspective of the information-theoretic channel in this survey. First, we presented the privacy threat model under information-theoretic channel. Second, we described and compared the information-theoretic channel models of GDP and LDP. Third, we summarized and analyzed definitions, privacy-utility metrics, properties, and mechanisms of GDP and LDP under their channel models. Finally, we discussed the open problems of GDP and LDP based on different types of information-theoretic channel models according to the above systematic review. Our main contribution provides a systematic survey of channel models, definitions, privacy-utility metrics, properties, and mechanisms for GDP and LDP from the perspective of information-theoretic channel and surveys the differential privacy synthetic data generation application using generative adversarial network and federated learning, respectively. Our work is helpful for systematically understanding the privacy threat model, definitions, privacy-utility metrics, properties, and mechanisms of GDP and LDP from the perspective of information-theoretic channel and promotes in-depth research and analysis of GDP and LDP based on different types of information-theoretic channel models.

show abstract

A Data Union Method Using Hierarchical Clustering and Set Unionability

Dang

Nguyen-Tan³

2021

Future Data and Security Engineering

View full text Add to dashboard Cite

Investigating Local Differential Privacy and Generative Adversarial Network in Collecting Data

Cited by 4 publications

References 22 publications

Inference attacks based on GAN in federated learning

Inference attacks based on GAN in federated learning

GDP vs. LDP: A Survey from the Perspective of Information-Theoretic Channel

A Data Union Method Using Hierarchical Clustering and Set Unionability

Contact Info

Product

Resources

About