Filters








14,729 Hits in 5.6 sec

GUPT

Prashanth Mohan, Abhradeep Thakurta, Elaine Shi, Dawn Song, David Culler
2012 Proceedings of the 2012 international conference on Management of Data - SIGMOD '12  
Unfortunately, it has seen limited adoption because of the loss in output accuracy, the difficulty in making programs differentially private, lack of mechanisms to describe the privacy budget in a programmer's  ...  Differential privacy provides a theoretical framework for processing data while protecting the privacy of individual records in a dataset.  ...  Acknowledgements We would like to thank Adam Smith, Daniel Kifer, Frank McSherry, Ganesh Ananthanarayanan, David Zats, Piyush Srivastava and the anonymous reviewers of SIGMOD for their insightful comments  ... 
doi:10.1145/2213836.2213876 dblp:conf/sigmod/MohanTSSC12 fatcat:nzohffxr5jb3xnmwwqappzbtoa

Utility Optimization of Federated Learning with Differential Privacy

Jianzhe Zhao, Keming Mao, Chenxi Huang, Yuyang Zeng, Shi Cheng
2021 Discrete Dynamics in Nature and Society  
To address the trade-off problems between privacy and utility in complex federated learning, a novel differentially private federated learning framework is proposed.  ...  Then, we design a multilevel and multiparticipant dynamic allocation method of privacy budget to reduce the injected noise, and the utility can be improved efficiently.  ...  privacy for the federated system and optimizes the utility by dynamic allocating differential privacy budgets, as shown in Figure 2 .  ... 
doi:10.1155/2021/3344862 fatcat:o67gabi7nnddllo7vju3avrmge

Combinational Randomized Response Mechanism for Unbalanced Multivariate Nominal Attributes

Xuejie Feng, Chiping Zhang, Jing Li, Linlin Dai
2020 IEEE Access  
Traditional local differential privacy algorithms usually assign the same privacy budget to all attributes, resulting in undesired frequency estimation.  ...  To obtain highly accurate of the results while satisfying local differential privacy, the aggregator needs to implement a reasonable privacy budget allocation scheme.  ...  task of local differential privacy, the service provider usually needs to determine the frequency distribution of each category in the user group.  ... 
doi:10.1109/access.2020.3013446 fatcat:ghzttva4ubftvbw5mrqpzrdcnm

A novel self-adaptive grid-partitioning noise optimization algorithm based on differential privacy

Zhaobin Liu, Haoze Lv, Minghui Li, Zhiyang Li, Zhiyi Huang
2019 Computer Science and Information Systems  
Of all the solutions of this problem, the differential privacy theory is based on strict mathematics and provides precise definition and quantitative assessed methods for privacy protection, it is widely  ...  To ensure that service providers can supply a completely optimal quality of service, users must provide exact location information. However, in that case, privacy disclosure accident is endless.  ...  This theorem indicates that the privacy protection level of combined algorithms is the sum of all budgets in a differential privacy protection model sequences [12] .  ... 
doi:10.2298/csis180901033l fatcat:bvohxruzbjggfla4uwv3j22v2e

Local differential privacy for unbalanced multivariate nominal attributes

Xuejie Feng, Chiping Zhang
2020 Human-Centric Computing and Information Sciences  
of satisfying local differential privacy.  ...  Then, we improved two popular local differential privacy mechanisms by taking advantage of the proposed privacy budget allocation techniques.  ...  Author' contributions XF designed and implemented the optimal allocation algorithm of the differential privacy budget and proved the effectiveness of the algorithm through experiments.  ... 
doi:10.1186/s13673-020-00233-x fatcat:nmn6txwzjva6beboaxlvnjdhra

Piezoelectric performance optimization of the PMN-PT based on self-adaptive differential evolution algorithm

Tundong Liu, Miao He, Zengruan Ye, Duanjun Lou, Sa Zhang, Qiao Sun
2017 Computational materials science  
Of all the solutions of this problem, the differential privacy theory is based on strict mathematics and provides precise definition and quantitative assessed methods for privacy protection, it is widely  ...  To ensure that service providers can supply a completely optimal quality of service, users must provide exact location information. However, in that case, privacy disclosure accident is endless.  ...  This theorem indicates that the privacy protection level of combined algorithms is the sum of all budgets in a differential privacy protection model sequences [12] .  ... 
doi:10.1016/j.commatsci.2016.08.046 fatcat:svkqaqjpjzeordk4mzsycsvb2m

On the Relationship Between Inference and Data Privacy in Decentralized IoT Networks [article]

Meng Sun, Wee Peng Tay
2019 arXiv   pre-print
information privacy and local differential privacy to within the predefined budgets.  ...  We propose an optimization framework in which both local differential privacy (data privacy) and information privacy (inference privacy) metrics are incorporated.  ...  This research is supported by the Singapore Ministry of Education Academic Research Fund Tier 1 grant 2017-T1-001-059 (RG20/17).  ... 
arXiv:1811.10322v3 fatcat:nqjqyqp35rfypfwcge3lwi3mei

Differentially Private Federated Learning for Resource-Constrained Internet of Things [article]

Rui Hu, Yuanxiong Guo, E. Paul. Ratazzi, Yanmin Gong
2020 arXiv   pre-print
differential privacy.  ...  The optimal schematic design of DP-PASGD that maximizes the learning performance while satisfying the limits on resource cost and privacy loss is formulated as an optimization problem, and an approximate  ...  In future work, we plan to study the performance of DP-PASGD in other learning settings such as multi-task learning and privacy considerations such as personalized differential privacy.  ... 
arXiv:2003.12705v1 fatcat:thczcju5obhrjduow6af4j45qy

PPDP-PCAO: An Efficient High-dimensional Data Releasing Method with Differential Privacy Protection

Wanjie LI, Xing ZHANG, Xiaohui LI, Guanghui CAO, Qingyun ZHANG
2019 IEEE Access  
PPDP-PCAO considers the existence of multi-sensitive attributes in highdimensional data, while the traditional methods of allocating privacy budgets cannot satisfy the personalized privacy protection.  ...  INDEX TERMS Differential privacy, evaluation mechanism, high-dimensional data, mutual information, principal component analysis optimization.  ...  In the case of satisfying differential privacy, we design a personalized allocation scheme of privacy budget, which makes the availability of the published data better.  ... 
doi:10.1109/access.2019.2957858 fatcat:4uy4og27irfdrnd6sqy3nnq66a

A Review of Differential Privacy in Individual Data Release

Jun Wang, Shubo Liu, Yongkai Li
2015 International Journal of Distributed Sensor Networks  
In this paper, the key aspects of basic concepts and implementation mechanisms related to differential privacy are explained, and the existing research results are concluded.  ...  Differential privacy is a relatively new notion of privacy and has become the de facto standard for a security-controlled privacy guarantee.  ...  is a near-optimal budgeting strategy.  ... 
doi:10.1155/2015/259682 fatcat:ieh3v7zgb5gqfnima3wsywcef4

Privacy-Preserving Monotonicity of Differential Privacy Mechanisms

Hai Liu, Zhenqiang Wu, Yihui Zhou, Changgen Peng, Feng Tian, Laifeng Lu
2018 Applied Sciences  
The trade-off of differential privacy shows that one thing increases and another decreases in terms of privacy metrics and utility metrics.  ...  In addition, we also theoretically and numerically analyzed the utility monotonicity of these several differential privacy mechanisms based on utility metrics of modulus of characteristic function and  ...  Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/app8112081 fatcat:ulrsk4znezbj5ayb4fyg5s6j54

A Workload Division Differential Privacy Algorithm to Improve the Accuracy for Linear Computations [chapter]

Jun Li, Huan Ma, Guangjun Wu, Yanqin Zhang, Bingnan Ma, Zhen Hui, Lei Zhang, Bingqing Zhu
2020 Lecture Notes in Computer Science  
In this paper, we propose a new simple ε-differential privacy algorithm.  ...  Differential privacy algorithm is an effective technology to protect data privacy, and there are many pieces of research about differential privacy and some practical applications from the Internet companies  ...  In this paper, we propose a novel ε-differential privacy algorithm, which uses Laplace-based noise and optimized workload division to decrease the computation error in complex data distribution for linear  ... 
doi:10.1007/978-3-030-50417-5_33 fatcat:cdbnyxwo4neixosyu6ltwhs3o4

DPCube: Differentially Private Histogram Release through Multidimensional Partitioning [article]

Yonghui Xiao, Li Xiong, Liyue Fan, Slawomir Goryczka
2012 arXiv   pre-print
Differential privacy is a strong notion for protecting individual privacy in privacy preserving data analysis or publishing.  ...  In this paper, we study the problem of differentially private histogram release for random workloads.  ...  Impact of Privacy Budget Allocation. We now study the impact of the allocation of the overall differential privacy budget α between the two phases.  ... 
arXiv:1202.5358v1 fatcat:yrlnlexdhrge3nlkpagnybjsd4

Three Tools for Practical Differential Privacy [article]

Koen Lennart van der Veen, Ruben Seggers, Peter Bloem, Giorgio Patrini
2018 arXiv   pre-print
the privacy budget, and ad-hoc privacy attacks are often required to test model privacy.  ...  bound to reduce the effective number of tuneable privacy parameters, and (3) we show that large-batch training improves model performance.  ...  As expected, in the centralized setting, we are able to learn the pattern without differential privacy but not with differential privacy. Distributed patterns should be learned.  ... 
arXiv:1812.02890v1 fatcat:tfq4gx56effzdd6ug37mnxuxqi

PLDP: Personalized Local Differential Privacy for Multidimensional Data Aggregation

Zixuan Shen, Zhihua Xia, Peipeng Yu, Liguo Zhang
2021 Security and Communication Networks  
To address it, local differential privacy (LDP) is proposed to protect the crowdsourced data without much loss of usage, which is popularly used in practice.  ...  However, the existing LDP protocols ignore users' personal privacy requirements in spite of offering good utility for multidimensional crowdsourced data.  ...  Differential Privacy. Differential privacy (DP) is a rigorous mathematical definition of privacy for securely sharing the statistic of a dataset on a server [7] .  ... 
doi:10.1155/2021/6684179 fatcat:vvirwwwwnfervlcfwnipjcnhyq
« Previous Showing results 1 — 15 out of 14,729 results