The growing prevalence of cloud computing and other forms of outsourced computation has made protecting computations executed over untrusted platforms or by untrusted agents a central focus of cryptography. Examples include secure distributed computation, secure database and memory access, secure delegation of computation, secure software distribution, and leakage- and tamper-resilient computation. Indeed, the need to harden or protect computations has become acute in the face of the proliferation of digital data and the growing need to outsource such data, its handling, and its monetization. In a world of outsourced data, we need to avoid the risk and/or policy violations that outsourcing trust to these outsourced providers would create.
Cryptographic schemes that address these issues have traditionally been developed in the circuit model of computation, and this model provides a good testing ground for the feasibility of solutions. However, in order to improve practical applicability, considerable recent effort has been devoted developing schemes that incur acceptable overhead even when applied to realistic computations and programs that are (as most programs are) designed for machines with random access memory.
This workshop will bring together cryptographers as well as security and programming language researchers to address the challenges of RAM-model cryptography, and to bridge the abstraction gap between cryptography and real-world programs. Topics of interest include schemes for oblivious memory access, oblivious algorithms, homomorphic encryption, program obfuscation, leakage resilient computation, functional encryption, as well as programming language techniques that automate RAM-model cryptography.
This workshop is a collaboration with the Modular Approach to Cloud Security project (MACS), an NSF Frontier project based at Boston University, and will be hosted by that project.