[PET] Special Issue on "Secure Privacy-preserving Machine Learning"

Takabi, Hassan Hassan.Takabi at unt.edu
Mon Jan 14 17:50:49 GMT 2019


*************************************************************************

Frontiers in Big Data Cybersecurity and Privacy

Special Issue on "Secure Privacy-preserving Machine Learning"

*************************************************************************


Machine learning algorithms, such as deep learning algorithms, have attracted attention as a breakthrough in computer vision, speech recognition, and other areas. However, machine learning algorithms require access to raw data which is often privacy sensitive. On the other hand, machine learning systems can be fragile and easily fooled by attacks that are effective across different architectures and applications. A novel combination of techniques is needed to enable secure privacy-preserving machine learning. Different approaches based on differential privacy-like techniques, homomorphic encryption where operations are performed on encrypted data, and techniques from the field of secure multi-party computation are being developed to deal with this problem.
The aim of this Article Collection is to bring together world-leading research on issues related to security and privacy of machine learning and data analytics and serve as a forum to unify different perspectives on this problem and explore the relative merits of each approach.
We invite submissions of novel research from academic researchers, practitioners and industry, both theory and application-oriented, on secure privacy-preserving machine learning and data analytics. The topics include but not limited to:
• Secure multi-party computation techniques for machine learning
• Homomorphic encryption techniques for machine learning
• Differential privacy techniques for machine learning
• Machine learning on encrypted data
• Adversarial machine learning (attacks and defenses)
• Hardware-based approaches to privacy-preserving machine learning
• Hardware-based approaches to learning on encrypted data
• Practical performance evaluations of different approaches to privacy-preserving machine learning
• Private multi-party machine learning
• Distributed privacy-preserving algorithms for private machine learning
• Decentralized protocols for learning on encrypted data
• Secure big data analytics
• Privacy-preserving big data analytics


IMPORTANT DATES

- Manuscript Due: February 4, 2019
- Preliminary Decisions: April 1, 2019
- Revised Manuscript: May 31, 2019
- Final Decisions: June 30, 2019
- Final Versions: July 15, 2019
- Publication Date: July 31, 2019


SUBMISSION GUIDELINES

https://www.frontiersin.org/research-topics/8734/secure-privacy-preserving-machine-learning
To help us with the review assignments, please submit an abstract if you plan to submit a paper.


EDITORS

Hassan Takabi

University of North Texas

Denton, TX, USA

takabi at unt.edu


Ran Gilad-Bachrach

Microsoft Research

rang at microsoft.com


Ebrahim Songhori

Google

esonghori at google.com



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.links.org/pipermail/pet/attachments/20190114/6c4802b0/attachment-0001.html>


More information about the PET mailing list