Filters








877 Hits in 4.8 sec

Data Security and Privacy in Cloud Computing: Concepts and Emerging Trends [article]

Rishabh Gupta, Deepika Saxena, Ashutosh Kumar Singh
2021 arXiv   pre-print
This paper envisages a discussion of cloud environment, its utilities, challenges, and emerging research trends confined to secure processing and sharing of data.  ...  Millions of users across the world leverages data processing and sharing benefits from cloud environment. Data security and privacy are inevitable requirement of cloud environment.  ...  [33] Multi-key fully Preserve privacy of Low efficiency homomorphic encryption sensitive data Ma et al.  ... 
arXiv:2108.09508v1 fatcat:4g57h5g3erdf5l4nnhfvpndc3u

Improving Utility of Differentially Private Mechanisms through Cryptography-based Technologies: a Survey [article]

Wen Huang, Shijie Zhou, Tianqing Zhu, Yongjian Liao
2021 arXiv   pre-print
Then, we summarize how to improve utility by combining differentially private mechanisms with homomorphic encryption schemes.  ...  Next, we summarize hardness results of what is impossible to achieve for differentially private mechanisms' utility from the view of cryptography.  ...  Bailey et al combine differential privacy mechanisms and fully homomorphic encryption schemes to tackle the problem of private set intersection [26] .  ... 
arXiv:2011.00976v2 fatcat:fizzcprz55cdxa7bwzyt7rzree

Secure Neuroimaging Analysis using Federated Learning with Homomorphic Encryption [article]

Dimitris Stripelis, Hamza Saleem, Tanmay Ghai, Nikhil Dhinagar, Umang Gupta, Chrysovalantis Anastasiou, Greg Ver Steeg, Srivatsan Ravi, Muhammad Naveed, Paul M. Thompson, Jose Luis Ambite
2021 arXiv   pre-print
In this work, we propose a framework for secure FL using fully-homomorphic encryption (FHE).  ...  Nevertheless, recent membership attacks show that private or sensitive personal data can sometimes be leaked or inferred when model parameters or summary statistics are shared with a central site, requiring  ...  CKKS In this work, we apply the Cheon-Kim-Kim-Song (CKKS) 34 fully-homomorphic construction, which is based on the hardness of the Learning-With-Error (LWE), 35 or its ring variant (RLWE 36 ) problem  ... 
arXiv:2108.03437v2 fatcat:5owpppcdrncq3a5xqoxb34nioq

Homomorphic proximity computation in geosocial networks

Peizhao Hu, Tamalika Mukherjee, Alagu Valliappan, Stanislaw Radziszowski
2016 2016 IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS)  
The underlying security is ensured by the homomorphic encryption scheme which supports computation on encrypted data.  ...  on encrypted location data.  ...  Since all functions can be broken down into these basic operations, we could theoretically construct Fully Homomorphic Encryption (FHE) schemes that perform arbitrary computations on encrypted data.  ... 
doi:10.1109/infcomw.2016.7562150 dblp:conf/infocom/HuMVR16 fatcat:5lttertrsvewnaalu322npm4be

Secure Federated Learning for Neuroimaging [article]

Dimitris Stripelis, Umang Gupta, Hamza Saleem, Nikhil Dhinagar, Tanmay Ghai, Rafael Sanchez, Chrysovalantis Anastasiou, Armaghan Asghar, Greg Ver Steeg, Srivatsan Ravi, Muhammad Naveed, Paul M. Thompson (+1 others)
2022 arXiv   pre-print
Second, neural parameters are encrypted before transmission and the community model is computed under fully-homomorphic encryption.  ...  Each site trains the neural network over its private data for some time, then shares the neural network parameters (i.e., weights, gradients) with a Federation Controller, which in turn aggregates the  ...  CKKS is fully homomorphic in that it supports an unbounded amount of arithmetic (or Boolean) circuit operations over encrypted data.  ... 
arXiv:2205.05249v1 fatcat:is5hz4psj5gd3ewznlesh7nypm

National strategy of digital cryptographic currency-digital bitcoin decentralized through the internet

Harold Szu
2018 MOJ Applied Bionics and Biomechanics  
Fully homomorphic cryptosystems have great practical implications in the outsourcing of private computations, for instance, in the context of cloud computing. 1 The problem of constructing a fully homomorphic  ...  Gentry's fully homomorphic encryption.  ...  with private sector client implementation.  ... 
doi:10.15406/mojabb.2018.02.00088 fatcat:yuwtwzjulbgezmmjauflk6w5qa

Outsourcing Medical Dataset Analysis: A Possible Solution [chapter]

Gabriel Kaptchuk, Matthew Green, Aviel Rubin
2017 Lecture Notes in Computer Science  
We implement one possible solution in which researchers operate directly on homomorphically encrypted data and the data owner decrypts the results.  ...  Specifically, we invoke differential privacy to protect information about individuals.  ...  (leveled) fully homomorphic encryption without bootstrapping. In S.  ... 
doi:10.1007/978-3-319-70972-7_6 fatcat:fksvsxwgqra4lj7dfx45kpaxre

Privacy-Preserving Machine Learning: Methods, Challenges and Directions [article]

Runhua Xu, Nathalie Baracaldo, James Joshi
2021 arXiv   pre-print
that increasingly restrict access to and use of privacy-sensitive data add significant challenges to fully benefiting from the power of ML for data-driven applications.  ...  Such a need for and the use of huge volumes of data raise serious privacy concerns because of the potential risks of leakage of highly privacy-sensitive information; further, the evolving regulatory environments  ...  According to the capabilities of performing various kinds of operations, typical HE types include partially homomorphic, somewhat homomorphic, leveled fully homomorphic, and fully homomorphic encryption  ... 
arXiv:2108.04417v2 fatcat:pmxmsbs2gvh6nd4jadcz4dnsrq

SARSA(0) Reinforcement Learning over Fully Homomorphic Encryption [article]

Jihoon Suh, Takashi Tanaka
2021 arXiv   pre-print
To achieve confidentiality, we implement computations over Fully Homomorphic Encryption (FHE).  ...  We then give a convergence result for the delayed updated rule of SARSA(0) with a blocking mechanism.  ...  fully homomorphic encryption, an efficient polynomial comparison function will be of significant value.  ... 
arXiv:2002.00506v2 fatcat:s55464glnzdlbankxjyocsk7om

PrivateGraph: A Cloud-Centric System for Spectral Analysis of Large Encrypted Graphs

Sagar Sharma, Keke Chen
2017 2017 IEEE 37th International Conference on Distributed Computing Systems (ICDCS)  
encrypted data, and the outcome of an important application of spectral analysis -spectral clustering.  ...  its audience interactively learn the major cloud-client interaction protocols: the privacypreserving data submission, the secure Lanczos and Nyström approximate eigen-decomposition algorithms that work over  ...  Two common generic privacy preserving approaches: fully homomorphic encryption (FHE) and secure multi-party garbled circuits (GC) [4] can theoretically construct the privacypreserving versions of most  ... 
doi:10.1109/icdcs.2017.189 dblp:conf/icdcs/SharmaC17 fatcat:nbnmyrpbxnbzbbp6sx2kwbjscq

A Survey on Secure Computation Based on Homomorphic Encryption in Vehicular Ad Hoc Networks

Xiaoqiang Sun, F. Richard Yu, Peng Zhang, Weixin Xie, Xiang Peng
2020 Sensors  
We first describe the related definitions and the current state of homomorphic encryption.  ...  Because homomorphic encryption supports computations of the ciphertext, it can completely solve this problem.  ...  Acknowledgments: We thank the reviewers for their detailed reviews and constructive comments, which have helped to greatly improve the quality of this paper.  ... 
doi:10.3390/s20154253 pmid:32751627 fatcat:2m2lw3vutbh4tbynvotylachye

MSCryptoNet: Multi-Scheme privacy-preserving deep learning in cloud computing

Owusu-Agyemeng Kwabena, Zhen Qin, Tianming Zhuang, Zhiguang Qin
2019 IEEE Access  
INDEX TERMS Internet of Things, privacy-preserving, fully homomorphic encryption.  ...  The MSCryptoNet is based on the multi-scheme fully homomorphic encryption.  ...  FULLY HOMOMORPHIC ENCRYPTION Fully homomorphic encryption (FHE) supports meaningful [19] unlimited addition and multiplication computations over encrypted data with results of addition and multiplications  ... 
doi:10.1109/access.2019.2901219 fatcat:gjlplklyfvgofkxlubd6crvgua

Deep Learning Security and Privacy

José Cabrero-Holgueras
2022 Zenodo  
The presentation was given to students of the Data Science Degree at UC3M.  ...  Homomorphic Encryption The RSA Homomorphism Privacy Preserving Deep Learning Membership Inference Attacks Data and Model Poisoning Attacks Use Differential Privacy Model Inversion Attacks Fully Homomorphic  ...  Input Data Security Training Fully Homomorphic Encryption Combining bootstrapping with the Learning With Errors Problem, you can theoretically make unbounded Output Data Security Training Differential  ... 
doi:10.5281/zenodo.6559344 fatcat:srkq73jhkbco3hwroaxrwf2jde

Privacy preserving distributed optimization using homomorphic encryption [article]

Yang Lu, Minghui Zhu
2018 arXiv   pre-print
For the first question, by using the techniques of homomorphic encryption, we propose novel algorithms which can achieve secure multiparty computation with perfect correctness.  ...  In particular, each participant holds a set of problem coefficients and/or states whose values are private to the data owner.  ...  Please refer to Section 9.1 for a summary of existing homomorphic encryption schemes.  ... 
arXiv:1805.00572v3 fatcat:oae2t5s77rcwbd6po3swjgoo7a

On the Black-box Use of Somewhat Homomorphic Encryption in NonInteractive Two-Party Protocols

Nirattaya Khamsemanan, Rafail Ostrovsky, William E. Skeith
2016 SIAM Journal on Discrete Mathematics  
Lastly, we also answer an open question from the thesis of Rappe [Rap06] regarding the construction of fully homomorphic encryption from group homomorphic encryption. * Abridged version appeared at CRYPTO  ...  Black-box usage of homomorphic encryption.  ...  constructing a fully homomorphic encryption scheme.  ... 
doi:10.1137/110858835 fatcat:nmkkqt27ubelngru5cdu3rvfpi
« Previous Showing results 1 — 15 out of 877 results