| NSF Org: |
CNS Division Of Computer and Network Systems |
| Recipient: |
|
| Initial Amendment Date: | August 18, 2016 |
| Latest Amendment Date: | August 18, 2016 |
| Award Number: | 1619261 |
| Award Instrument: | Standard Grant |
| Program Manager: |
Daniela Oliveira
doliveir@nsf.gov (703)292-0000 CNS Division Of Computer and Network Systems CSE Directorate for Computer and Information Science and Engineering |
| Start Date: | October 1, 2016 |
| End Date: | September 30, 2020 (Estimated) |
| Total Intended Award Amount: | $298,231.00 |
| Total Awarded Amount to Date: | $298,231.00 |
| Funds Obligated to Date: |
|
| History of Investigator: |
|
| Recipient Sponsored Research Office: |
9500 GILMAN DR LA JOLLA CA US 92093-0021 (858)534-4896 |
| Sponsor Congressional District: |
|
| Primary Place of Performance: |
9500 Gilman Dr La Jolla CA US 92093-0407 |
| Primary Place of
Performance Congressional District: |
|
| Unique Entity Identifier (UEI): |
|
| Parent UEI: |
|
| NSF Program(s): | Secure &Trustworthy Cyberspace |
| Primary Program Source: |
|
| Program Reference Code(s): |
|
| Program Element Code(s): |
|
| Award Agency Code: | 4900 |
| Fund Agency Code: | 4900 |
| Assistance Listing Number(s): | 47.070 |
ABSTRACT
Computing on sensitive data is a standing challenge central to several modern-world applications. Secure Function Evaluation (SFE) allows mistrusting parties to jointly compute an arbitrary function on their private inputs without revealing anything but the result. The GC@Scale project focuses on novel scalable methods for addressing SFE, which directly translate to stronger cryptography and security for myriads of tasks with sensitive data. The applications are wide reaching and include privacy-preserving processing of medical, genome, and biometric data, as well as personal, government, and industrial cloud computing. The project includes an ambitious educational program that targets both undergraduate/ graduate students, and also addresses issues related to outreach.
The concept of SFE using Garbled Circuits (GC) was introduced by Yao. Despite a decade of research in GC implementation and several key progresses, scalability of the available methods has been hampered by the circuit representation as a directed acyclic graph, and software-level local logic optimizations. GC@Scale leverages PI's recent work, which has changed the SFE landscape by viewing GC generation as an atypical sequential logic synthesis. The project plans to advance the understanding and enable expanded exploration of SFE methodologies, while simultaneously enriching the theory, practice, and tools for logic design, synthesis, mapping and optimization. The proposed plan includes: (i) design and FPGA implementation of an efficient general purpose Garbled Processor for secure computation; (ii) Creating the challenging application-specific GC matching and search engines with a higher than linear complexity. (iii) Devising new custom SFE engines for Machine Learning tasks.
PUBLICATIONS PRODUCED AS A RESULT OF THIS RESEARCH
Note:
When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external
site maintained by the publisher. Some full text articles may not yet be available without a
charge during the embargo (administrative interval).
Some links on this page may take you to non-federal websites. Their policies may differ from
this site.
PROJECT OUTCOMES REPORT
Disclaimer
This Project Outcomes Report for the General Public is displayed verbatim as submitted by the Principal Investigator (PI) for this award. Any opinions, findings, and conclusions or recommendations expressed in this Report are those of the PI and do not necessarily reflect the views of the National Science Foundation; NSF has not approved or endorsed its content.
Security and privacy are paramount concerns in development and employment of technologies and applications. The results of this project are of interest to fabless semiconductor companies, US government agencies and university researchers worldwide. Benefits to the society include secure and trustworthy electronics for healthcare, defense, finance, transportation, automotive and other applications. The results will be disseminated in technical events, via Trust-Hub website and in open-source format.
This project is impacting the field in myriads of ways, starting from the development of basic knowledge, theory, research, all the way to publications, open-source software, and open-source teaching material that all contribute to the healthy and constant progress of the field.
In particular, the methods for efficient machine learning are based on fundamental newly developed methodologies that integrate a combination of hardware characterization/computer engineering, which aim to significantly improve performance.
Our research on hardware acceleration of secure machine learning has accelerated the privacy-preserving inference by 3 orders of magnitude. Data privacy is one of the most discussed and debated topics in the past few years. Still ensuring complete privacy of the user data had been considered close to impossible. Our work brings the complexity of this task within practical limits and shows the promise to be adopted in the industry within few years.
The project has made a significant mark in the development of the field, by presenting the following results in the last year of the review period:
* MPCircuits, the first scalable multi-party security platform.
* ARM2GC, a new compiler that translates the commands from a high level (ARM c++ compiled code) to GC framework.
* Chameleon, a new hybrid protocol for privacy-preserving deep learning which reported the most efficient results to date (on the day of publication) in the field.
* FASE, a general-purpose FPGA framework for acceleration of the GC.
The material developed in this project is being used in PI's graduate hardware and embedded security course in UCSD. The PI is teaching an undergraduate course in FPGA Verilog. The hands-on project for that course draws upon the material devised in this project and is about bitcoin hashing on FPGA.
The PI has actively helped with organizing workshops and conferences or contributed tutorials and invited sessions to these event. The tools and methodology devised in this project are used and cited by other researchers in the field (e.g., the Chamelon paper alone is 125 times cited) and by myriads of researchers and practitioner who are interested in privacy-preserving applications.
Last Modified: 11/15/2021
Modified by: Farinaz Koushanfar
Please report errors in award information by writing to: awardsearch@nsf.gov.
