An effective and practical gradient inversion attack

Luo, Zeren; Zhu, Chuangwei; Fang, Lujie; Kou, Guang; Hou, Ruitao; Wang, Xianmin

doi:10.1002/int.22997

Cited by 3 publications

(2 citation statements)

References 35 publications

(88 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For FL, Zhu et al [174] demonstrated how can we obtain the training data from shared gradients by developing a Deep Leakage from the gradients. Luo et al [85] presented another idea of gradient inversion by reconstructing training data. Using transfer learning, Ye et al [159] proved that the inversion attacks fall apart when we target the student model.…”

Section: Model Inversion Attackmentioning

confidence: 99%

Digital Privacy Under Attack: Challenges and Enablers

Song¹,

Deng²,

Pokhrel³

et al. 2023

Preprint

View full text Add to dashboard Cite

Users have renewed interest in protecting their private data in the digital space. When they don't believe that their privacy is sufficiently covered by one platform, they will readily switch to another. Such an increasing level of privacy awareness has made privacy preservation an essential research topic. Nevertheless, new privacy attacks are emerging day by day. Therefore, a holistic survey to compare the discovered techniques on attacks over privacy preservation and their mitigation schemes is essential in the literature. We develop a study to fill this gap by assessing the resilience of privacy-preserving methods to various attacks and conducting a comprehensive review of countermeasures from a broader perspective. First, we introduce the fundamental concepts and critical components of privacy attacks. Second, we comprehensively cover major privacy attacks targeted at anonymous data, statistical aggregate data, and privacy-preserving models. We also summarize popular countermeasures to mitigate these attacks. Finally, some promising future research directions and related issues in the privacy community are envisaged. We believe this survey will successfully shed some light on privacy research and encourage researchers to entirely understand the resilience of different existing privacy-preserving approaches.

show abstract

Section: Model Inversion Attackmentioning

confidence: 99%

Digital Privacy Under Attack: Challenges and Enablers

Song¹,

Deng²,

Pokhrel³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Te virtual data are learned using an optimization algorithm in such a way that the gradient obtained by backpropagation on the common model is similar to the real gradient, and the training data and labels are obtained after several rounds of iterative optimization. At the moment, this is one of the hottest topics in the study of variants of DLG-based methods [9,[52][53][54].…”

Section: Gradient Update-based Data Leakagementioning

confidence: 99%

Fast and Accurate Deep Leakage from Gradients Based on Wasserstein Distance

Peng

Tan

2023

International Journal of Intelligent Systems

View full text Add to dashboard Cite

Shared gradients are widely used to protect the private information of training data in distributed machine learning systems. However, Deep Leakage from Gradients (DLG) research has found that private training data can be recovered from shared gradients. The DLG method still has some issues such as the “Exploding Gradient,” low attack success rate, and low fidelity of recovered data. In this study, a Wasserstein DLG method, named WDLG, is proposed; the theoretical analysis shows that under the premise that the output layer of the model has a “bias” term, predicting the “label” of the data by whether the “bias” is “negative” or not is independent of the approximation of the shared gradient, and thus, the label of the data can be recovered with 100% accuracy. In the proposed method, the Wasserstein distance is used to calculate the error loss between the shared gradient and the virtual gradient, which improves model training stability, solves the “Exploding Gradient” phenomenon, and improves the fidelity of the recovered data. Moreover, a large learning rate strategy is designed to improve model training convergence speed in-depth. Finally, the WDLG method is validated on datasets from MNIST, Fashion MNIST, SVHN, CIFAR-100, and LFW. Experiments results show that the proposed WDLG method provides more stable updates for virtual data, a higher attack success rate, faster model convergence, higher image fidelity during recovery, and support for designing large learning rate strategies.

show abstract

Towards Risk-Free Trustworthy Artificial Intelligence: Significance and Requirements

Alzubaidi,

Al-Sabaawi,

Bai

et al. 2023

International Journal of Intelligent Systems

View full text Add to dashboard Cite

Given the tremendous potential and influence of artificial intelligence (AI) and algorithmic decision-making (DM), these systems have found wide-ranging applications across diverse fields, including education, business, healthcare industries, government, and justice sectors. While AI and DM offer significant benefits, they also carry the risk of unfavourable outcomes for users and society. As a result, ensuring the safety, reliability, and trustworthiness of these systems becomes crucial. This article aims to provide a comprehensive review of the synergy between AI and DM, focussing on the importance of trustworthiness. The review addresses the following four key questions, guiding readers towards a deeper understanding of this topic: (i) why do we need trustworthy AI? (ii) what are the requirements for trustworthy AI? In line with this second question, the key requirements that establish the trustworthiness of these systems have been explained, including explainability, accountability, robustness, fairness, acceptance of AI, privacy, accuracy, reproducibility, and human agency, and oversight. (iii) how can we have trustworthy data? and (iv) what are the priorities in terms of trustworthy requirements for challenging applications? Regarding this last question, six different applications have been discussed, including trustworthy AI in education, environmental science, 5G-based IoT networks, robotics for architecture, engineering and construction, financial technology, and healthcare. The review emphasises the need to address trustworthiness in AI systems before their deployment in order to achieve the AI goal for good. An example is provided that demonstrates how trustworthy AI can be employed to eliminate bias in human resources management systems. The insights and recommendations presented in this paper will serve as a valuable guide for AI researchers seeking to achieve trustworthiness in their applications.

show abstract

An effective and practical gradient inversion attack

Cited by 3 publications

References 35 publications

Digital Privacy Under Attack: Challenges and Enablers

Digital Privacy Under Attack: Challenges and Enablers

Fast and Accurate Deep Leakage from Gradients Based on Wasserstein Distance

Towards Risk-Free Trustworthy Artificial Intelligence: Significance and Requirements

Contact Info

Product

Resources

About