Differential Privacy Meets Multi-Party Computation (DPMPC) Workshop
Monday & Tuesday, June 4 – 5, 2018
Rajen Kilachand Center for Integrated Life Sciences & Engineering
610 Commonwealth Ave
Boston, MA 02215
Multi Party Secure Computation (MPC) protocols originating with work in the eighties enable multiple parties to run joint computations on their individual secret inputs so that nothing except for the output of the computation is revealed to each other, in the presence of various adversarial models. MPC are particularly relevant today when the data necessary to make discoveries often involves massive data sets that are held by different parties – where these parties need to keep their data private due to regulation, or wish to monetize their data rather than share it freely. Differential Private (DP) algorithms address the problem of how to analyze data in a privacy-preserving manner. Here the question is how to query a database of sensitive information, so that individual entries are not exposed and yet meaningful and robust data analysis can be performed.
Both MPC and DP address related but different privacy concerns. Where one technology applies the other one may not, and sometime both technologies should be used to gain the security necessary. This workshop is dedicated to understanding both technologies, when they apply, and what is the possible synergy between them.
Shafi Goldwasser (MIT, Berkeley, Weizmann), Ran Canetti (BU, TAU)
Online registration is requested (registration is free).
Monday, June 4
|8:30 am – 9:15 am||Breakfast|
|9:15 am – 9:30 am||Opening Remarks|
|9:30 am – 10:30 am||
Salil Vadhan (Harvard): Multiparty Differential Privacy
|10:30 am – 10:45 am||Break|
|10:45 am – 11:30 am||
Anthony Philippakis (Broad): Privacy in Genomics research
|11:30 am – 11:50 am||
Elaine Shi (Cornell): Foundations of Differentially Oblivious Algorithms
Inspired by the elegant notion of differential privacy, we initiate the study of a new notion of access pattern privacy, which we call “(eps, delta)-differential obliviousness”. We separate the notion of (eps, delta)-differential obliviousness from classical obliviousness by considering several fundamental algorithmic abstractions including sorting small-length keys, merging two sorted lists, and range query data structures (akin to binary search trees). We show that by adopting differential obliviousness with reasonable choices of eps and delta, not only can one circumvent several impossibilities pertaining to full obliviousness, one can also, in several cases, obtain meaningful privacy with little overhead relative to the non-private baselines (i.e., having privacy “almost for free”). On the other hand, we show that for very demanding choices of eps and delta, the same lower bounds for oblivious algorithms would be preserved for (eps, delta)-differential obliviousness.
|11:50 am – 12:05 pm||
Joshua Baron (DARPA): DARPA efforts in MPC
Abstract & presentation private per presenter’s request.
|12:05 pm – 1:00 pm||Lunch|
|1:00 pm – 1:30 pm||
Aloni Cohen (MIT): Anonymization, Singling Out, and the GDPR
A host of data privacy laws govern in the US and abroad govern the use of certain data (e.g., of clients, users, patients, students, subjects, children). While cryptographic tools might appear to provide easy solutions for compliance, a significant conceptual gap makes it difficult to draw strong conclusions.
Effective May 25, the GDPR restricts the use of “personal data.” Using the GDPR and related texts as a foundation, we mathematically model one facet of personal data — its susceptibility to “singling out.” With a precise definition in hand, we study a variety of “anonymization” techniques with the goal of drawing both mathematical and legal conclusions.
|1:30 pm – 2:30 pm||Panel: DP and MPC in the eyes of the law
Panelists: Ahmed Ghappour (BU Law), Paul Ohm (Georgetown Law), Salil Vadhan (Harvard), Ran Canetti (BU)
Moderator: Shafi Goldwasser
|2:30 pm – 3:00 pm||Coffee|
|3:00 pm – 3:45 pm||
Xi He (Duke): Trading off Accuracy Privacy and Efficiency in Secure Joins using Differential Privacy
|3:45 pm – 4:25 pm|
|4:25 pm – 4:55 pm||
Ran Cohen (MIT & NEU): Applying Modern Crypto to Advance Social Science
In this talk I will present a collaboration with the Institute for Research on Innovation and Science (IRIS). We demonstrate how to provide strong privacy guarantees by combining fully homomorphic encryption (FHE) and secure multiparty computation (MPC) to carry out regression analysis over hidden data.
Tuesday, June 5
|8:30 am – 9:00 am||Breakfast|
|9:00 am – 9:45 am|
|9:45 am – 10:30 am||
Anand Sarwate (Rutgers): Jana: Private Data as a Service
Joint work with David Archer (Galois), Nigel Smart (Bristol), Rebecca Wright (Rutgers), David Cash (Rutgers), and Dov Gordon (George Mason).
|10:30 am – 11:00 am||Break|
|11:00 am – 11:30 am||
Sahar Mazloom (George Mason):Differentially Private Access Patterns in Secure Computation
We then demonstrate that this leakage is useful in a broad class of computations. We show that computations such as histograms, PageRank and matrix factorization, which can be performed in common graph-parallel frameworks such as MapReduce or Pregel, benefit from our relaxation. We implement a protocol for securely executing graph-parallel computations, and evaluate the performance on the three examples just mentioned above. We demonstrate marked improvement over prior implementations for these computations.
|11:30 am – 12:00 pm||
Eyal Ronen (Weizmann): How to (not) share a password: Privacy preserving protocols for finding heavy hitters with adversarial behavior
Ronen will present a method to help protect the Internet from such large scale attacks. He enables a server to identify popular passwords (heavy hitters), and publish a list of over-popular passwords that must be avoided. This filter ensures that no single password can be used to compromise a large percentage of the users. The list is dynamic and can be changed as new users are added or when current users change their passwords. He applies maliciously secure two-party computation and differential privacy to protect the users’ password privacy. His solution does not require extra hardware or cost, and is transparent to the user.
His private heavy hitters construction is secure even against a malicious coalition of devices which tries to manipulate the protocol in order to hide the popularity of some password that the attacker is exploiting. It also ensures differential privacy under continues observation of the blacklist as it changes over time.
As a reality check we verified the following three tests: computed the guarantees that the system provides with respect to a few publicly available data points, ran simulations on various distributions, and implemented and analyzed a proof-of-concept on an IoT device.
His construction can also be used in other settings to privately learn heavy hitters in the presence of an active malicious adversary. E.g., learning the most popular sites accessed by the TOR network.
|12:00 pm – 1:00 pm||Lunch|
|1:00 pm – 1:45 pm|
|1:45 pm – 2:15 pm||
Andrei Lapets (BU): JIFF: Platform for MPC Protocol Prototyping and Privacy-Preserving Web Application Development
|2:15 pm – 2:30 pm||Break|
|2:30 pm – 3:00 pm||
Ariel Nof (BIU): Fast Large-Scale Honest-Majority MPC for Malicious Adversaries
In this paper, Nof presents new protocols for securely computing any functionality represented by an arithmetic circuit. Nof utilizes a new method for verifying that the adversary does not cheat, that yields a cost of just twice that of semi-honest protocols in some settings. The protocols are information-theoretically secure in the presence of a malicious adversaries, assuming an honest majority. Nof presents protocol variants for small and large fields, and shows how to efficiently instantiate them based on replicated secret sharing and Shamir sharing. As with previous works in this area aiming to achieve high efficiency, the protocol is secure with abort and does not achieve fairness, meaning that the adversary may receive output while the honest parties do not.
The protocol was implemented and ran in experiments for different numbers of parties, different network configurations and different circuit depths. The protocol significantly outperforms the previous best for this setting; for a large number of parties, this implementation runs almost an order of magnitude faster than theirs.
|3:00 pm – 3:45 pm|
Boston University is offering shared suites with a private bedroom and a shared bathroom at 10 Buick Street. The cost is $90 per person, per night, available from June 3 through the night of June 6. For more information, visit the Event Accommodations website.
Complimentary parking is available for those who will be driving. Please contact Katherine D’Angelo at firstname.lastname@example.org, and include your first and last name and what days you will be needing parking. The deadline to request a parking pass is Friday, June 1st at 3:00 pm.
Founded in 1934 by industrialist Alfred P. Sloan Jr., the Foundation is a not-for-profit grantmaking institution that supports high quality, impartial scientific research; fosters a robust, diverse scientific workforce; strengthens public understanding and engagement with science; and promotes the health of the institutions of scientific endeavor.
The Modular Approach to Cloud Security (MACS) project is a $10 million five-year National Science Foundation’s (NSF) Secure and Trustworthy Cyberspace (SaTC) program Frontier grant awarded to Boston University in 2014. MACS is one of two new center-scale “Frontier” awards to support large, multi-institution projects that address grand challenges in cybersecurity science and engineering with the potential for broad economic and scientific impact.
For more information, please contact Ran Canetti, Professor of Computer Science & Director of Center of Reliable Information System and Cyber Security, at email@example.com.