Computing That Serves

Privacy and Secure Multi-party Computation


Thursday, September 15, 2016 - 11:00am


Michael Clark


Kent Seamons

Michael Clark
Thursday, September 15, 2016
11:00amĀ  1170 TMCB

Digitization of various aspects of our lives has grown rapidly in recent years. Industries ranging from public utilities to medicine to finance have benefited greatly from collecting and analyzing private, personal information. However, privacy concerns have been raised by researchers and thought leaders surrounding the types of data being collected. This has lead to costly data protection regulations and a decrease in adoption of certain technologies. Additionally, with data breaches on the rise, it is equally concerning what hackers with unauthorized access to disparate data sources could discover by combining such information. In this talk, I discuss privacy in general and some of the specific concerns that have been raised. I describe one solution to the problem, a cryptographic technique known as secure multiparty computation (MPC), that allows distrusting parties with private data to jointly compute on their private data without having to reveal their data publicly. Instead, they compute on encrypted versions of the data and only reveal the output of the computation. I then present an optimization to MPC that enables distributed, secure computation, called transferable secure multiparty computation (T-MPC). T-MPC changes the adversary model slightly for privacy-preserving computations but enables orders of magnitude more efficient applications. I describe one such application, smart grid optimization via smart metering and show comparisons between using MPC protocols and T-MPC protocols.


Dr. Michael Clark is a Research Scientist with Tenet 3, LLC, a cyber security company in the Dayton, OH area. His research interests include cryptographic methods for privacy-preserving computation, hardware and application level cryptanalysis, security metrics, and the convergence of big data and cyber security problems. He received a PhD in Computer Science from the Air Force Institute of Technology (AFIT), at Wright-Patterson Air Force Base, in 2015. His research at AFIT focused on optimizing secure multi-party computations for distributed processing and invented transferable secure multiparty computation (T-MPC) in the process. In 2010, he received a M.S. in Computer Science from the University of Utah, where he studied collaborative methods for optimizing covert communications. He received a B.S. in Computer Science from Brigham Young University in 2008. Dr. Clark has served as a moderator for Stack Exchange's Cryptography site ( for the last four years. In this role he has reached out to hundreds of thousands of people interested in cryptography, ranging in expertise from academic researchers to hobbyists.