Search results
1 – 2 of 2Yong Ding, Peixiong Huang, Hai Liang, Fang Yuan and Huiyong Wang
Recently, deep learning (DL) has been widely applied in various aspects of human endeavors. However, studies have shown that DL models may also be a primary cause of data leakage…
Abstract
Purpose
Recently, deep learning (DL) has been widely applied in various aspects of human endeavors. However, studies have shown that DL models may also be a primary cause of data leakage, which raises new data privacy concerns. Membership inference attacks (MIAs) are prominent threats to user privacy from DL model training data, as attackers investigate whether specific data samples exist in the training data of a target model. Therefore, the aim of this study is to develop a method for defending against MIAs and protecting data privacy.
Design/methodology/approach
One possible solution is to propose an MIA defense method that involves adjusting the model’s output by mapping the output to a distribution with equal probability density. This approach effectively preserves the accuracy of classification predictions while simultaneously preventing attackers from identifying the training data.
Findings
Experiments demonstrate that the proposed defense method is effective in reducing the classification accuracy of MIAs to below 50%. Because MIAs are viewed as a binary classification model, the proposed method effectively prevents privacy leakage and improves data privacy protection.
Research limitations/implications
The method is only designed to defend against MIA in black-box classification models.
Originality/value
The proposed MIA defense method is effective and has a low cost. Therefore, the method enables us to protect data privacy without incurring significant additional expenses.
Details
Keywords
Junsong Jia, Zhihai Gong, Chundi Chen, Huiyong Jian and Dongming Xie
This paper aims to provide a typical example of accounting for the carbon dioxide equivalent (CO2e) in underdeveloped cities, especially for the Poyang Lake area in China. The…
Abstract
Purpose
This paper aims to provide a typical example of accounting for the carbon dioxide equivalent (CO2e) in underdeveloped cities, especially for the Poyang Lake area in China. The accounting can increase public understanding and trust in climate mitigation strategies by showing more detailed data.
Design/methodology/approach
The paper uses the “Global Protocol for Community-scale greenhouse gas emission inventories (GPC)” method, a worldwide comparable framework for calculating urban CO2e emission (CE). The empirical case is an underdeveloped city, Nanchang, in China.
Findings
The results show the total CE of Nanchang, containing the electricity CE of Scope 2, grew rapidly from 12.49 Mt in 1994 to 55.00 Mt in 2014, with the only recession caused by the global financial crisis in 2008. The biggest three contributors were industrial energy consumption, transportation and industrial processes, which contributed 44.71-72.06, 4.10-25.07 and 9.07-22.28 per cent, respectively, to the total CE. Almost always, more than 74.41 per cent of Nanchang’s CE was related to coal. When considering only the CEs from coal, oil and gas, these CEs per unit area of Nanchang were always greater than those of China and the world. Similarly, these CEs per gross domestic product of Nanchang were always bigger than those of the world. Thus, based on these conclusions, some specific countermeasures were recommended.
Originality/value
This paper argues that the CO2e accounting of underdeveloped cities by using the GPC framework should be promoted when designing climate mitigation policies. They can provide more scientific data to justify related countermeasures.
Details