We design Synq, a system that addresses privacy concerns by supporting public policy analytics over encrypted data. We specifically use an application-centric approach which drives Synq’s design around a study conducted on the opioid epidemic in Massachusetts. We systematize the design considerations of the public policy context and demonstrate how the combination of design considerations that Synq addresses is novel through a survey of the literature. We then present our protocol which combines structured encryption, somewhat homomorphic encryption, and oblivious pseudorandom functions to support a complex query language that includes filtering (retrieving rows by attribute/value pairs), linking (merging rows from different tables that represent the same individual) and aggregate functions (sum, count, average, variance, regression). We formally express the security of our protocol and show that Synq is efficient in practice while satisfying more demanding constraints than prior work.

We present the first systematic study of multi-attribute range search on a symmetrically encrypted database outsourced to an honest-but-curious server. Prior work includes a thorough analysis of single-attribute range search schemes (e.g. Demertzis et al. 2016) and a proposed high-level approach for multi-attribute schemes (De Capitani di Vimercati et al. 2021). We first introduce a flexible framework for building secure range search schemes with an arbitrary number of attributes (dimensions) by adapting a broad class of geometric search data structures to operate on encrypted data. Our framework encompasses widely used data structures such as multi-dimensional range trees and quadtrees, and has strong security properties that we formally prove. We then develop six concrete highly parallelizable range search schemes within our framework that offer a sliding scale of efficiency and security tradeoffs to suit the needs of the application. We evaluate our schemes with a formal complexity and security analysis, a prototype implementation, and an experimental evaluation on real-world datasets.

We present the first database reconstruction attacks against response-hiding private range search schemes on encrypted databases of arbitrary dimensions. Falzon et al. (VLDB 2022) present a number of range-supporting schemes on arbitrary dimensions exhibiting different security and efficiency trade-offs. Additionally, they characterize a form of leakage, structure pattern leakage, also present in many one-dimensional schemes e.g., Demertzis et al. (SIGMOD 2016) and Faber et al. (ESORICS 2015). We present the first systematic study of this leakage and attack a broad collection of schemes, including schemes that allow the responses to contain false-positives (often considered the gold standard in security). We characterize the information theoretic limitations of a passive persistent adversary. Our work shows that for range queries, structure pattern leakage can be as vulnerable to attacks as access pattern leakage. We give a comprehensive evaluation of our attacks with a complexity analysis, a prototype implementation, and an experimental assessment on real-world datasets.

We develop ARQ, a systematic framework for creating cryptographic schemes that handle range aggregate queries (sum, minimum, median, and mode) over encrypted datasets. ARQ unifies structures from the plaintext data management community with existing STE primitives. We prove how such combinations yield efficient (and secure) constructions in the encrypted setting. As part of this work, we designed and implemented a new, open-source, encrypted search library called Arca and implemented the ARQ framework using this library in order to evaluate ARQ’s practicality. Our experiments on real-world datasets demonstrate the efficiency of the schemes derived from the ARQ framework in comparison to prior work.

Software exploitation techniques and state-of-the-art mechanisms for hardening software. With Vasileios Kemerlis.

An introduction to principles of computer security from an applied viewpoint and provides hands-on experience on security threats and countermeasures. Topics include cryptosystems, web security, network security, malware, code execution vulnerabilities, access control, cryptocurrencies, machine learning, and human and social issues. With Roberto Tamassia (2019, 2020) and Bernardo Palazzi (2021).

• 2021 contributions:
  • Interviewed, hired, trained, and coordinated staff of 10 undergraduate and graduate TAs.
  • Recurrent guest lecturer in 9 lectures during the semester; delivered over 4 hours of lecture content during the semester.
  • Co-developed and hosted 8 “sections” covering supplementary course content during the semester; developed “discussion” handouts for the sections to help foster community and student-to-student interaction in a remote learning environment.
  • Developed new course project on cloud storage security and secure systems design. Co-designed and developed an implementation-agnostic, adversarial autograder for evaluating and discovering attacks against student implementations, reducing grading workload from 100+ hours of combined TA work hours to ~10 hours of combined TA work hours.
  • Coordinated homework and exam design with staff and professor. Wrote ~50% of the exam problems for 2021.
• 2020 contributions:
  • Interviewed, hired, trained, and coordinated staff of 11 undergraduate and graduate TAs.
  • Organized creation of new homeworks with a more open-ended, design-based focus; modernized homework content (with questions on cryptography, MPC, web security, networks, data compression, anonymization networks, etc.) and wrote +30 new questions for the entire semester with additional “reserve questions” for future years.
  • Designed new “midterm” component of course focused on open-ended design and security analysis questions; wrote ~80% of the exam problems for 2020.
  • Redeveloped the web security and operating security technical projects to improve learning outcomes and increase efficiency of internal grading processes; facilitiated technical project logistics.
  • Created UTA Manual outlining internal staff processes and responsibilities.
• 2019 contributions:
  • Interviewed, hired, trained, and coordinated staff of 8 undergraduate and graduate TAs.
  • Ported technical components of projects written in Bash, PHP, Javascript, and Go to Google Cloud Platform’s Compute Engine.
  • Automated setup process of sandbox Linux virtual machines for each project using the Compute Engine API, reducing setup times by up to ~92%.
  • Co-gave “Passwords” lecture with hashcat password recovery demonstration (slide deck, video starting @ 33:27).

Explores the principles of modern programming languages by implementation; studies data and their types, including polymorphism, type inference, and type soundness; examines compiler and run-time system topics: continuation-passing style and garbage collection. With Shriram Krishnamurthi.

• 2020 contributions:
  • Interviewed and hired staff of 6 undergraduate TAs.
  • Heavily rewrote 9 assignment specifications on programming language feature implementations in Plait and Racket (examples include “Interpreter”, “Type Checker”, “Type Inference”, “Generators”).
  • Transitioned internal staff infrastructure to Gradescope to automate the majority of grading and feedback distribution; developed autograder framework to emulate a Racket version of Examplar, a research IDE helping students assess their understanding of specifications via example-driven development.
• 2019 contributions:
  • Interviewed and hired staff of 6 undergraduate TAs.
  • Remotely answered student-submitted questions throughout the semester.
  • Copy-edited several assignment specifications written by other TAs.

Functional programming, data structures, and algorithms in Racket and Pyret. Includes a summer component taught using the first half of How to Design Programs, then transitions to content based on portions of Programming and Programming Languages during the semester. With Shriram Krishnamurthi.

• Interviewed, hired, trained, and coordinated staff of 9 undergraduate TAs.
• Remotely organized summer placement process and grading for 174 students.
• Rewrote old assignments (“Tour Guide”) and developed new labs (“Iterating Over Trees”, “Tensorflow”).
• Maintained and added new features to existing, end-to-end Google Apps Script-based pipeline for grading management and distribution.
• Wrote internal handbook for future iterations of course staff, covering suggestions for recruiting and hiring future teaching assistants, organizing course logistics, running existing course scripts, and more.

Introduction to programming in MATLAB and Python, with an emphasis on STEM data analysis and simulation problems. With Dan Potter.

Data-focused introduction to computer science using Pyret. With Kathi Fisler.