Edo Roth edoroth@duck.com
My face

Hello and welcome! I am a privacy researcher and engineer, and I currently work as a software engineer on a federated analytics research team at Google. I received my PhD from the Department of Computer and Information Science at the University of Pennsylvania in 2022. I was fortunate to be advised by Andreas Haeberlen, and I also spent two great summers interning at Microsoft Research.

My past research has focused on privacy and security in data analytics systems, particularly in large-scale distributed ones. I have worked on differential privacy, applied cryptography, federated learning, and distributed systems.

In general, I'm motivated to work on public-interest technology and studying/helping to build trustworthy and healthy digital infrastructure.

I grew up in Champaign-Urbana, Illinois, and completed my bachelor studies in Computer Science at Columbia University.

Research

2025

Is API Access to LLMs Useful for Generating Private Synthetic Tabular Data?

Marika Swanberg, Ryan McKenna, Edo Roth, Albert Cheu, Peter Kairouz
[PDF]

Mayfly: Private Aggregate Insights from Ephemeral Streams of On-Device User Data

Christopher Bian, Albert Cheu, Stanislav Chiknavaryan, Zoe Gong, Marco Gruteser, Oliver Guinan, Yannis Guzman, Peter Kairouz, Artem Lagzdin, Ryan McKenna, Grace Ni, Edo Roth, Maya Spivak, Timon Van Overveldt, Ren Yi
[PDF]

2024

Releasing Large-Scale Human Mobility Histograms with Differential Privacy

Christopher Bian, Albert Cheu, Yannis Guzman, Marco Gruteser, Peter Kairouz, Ryan McKenna, Edo Roth*.
*authors in alphabetical order, was one of two primary contributors.
[PDF]

2023

Arboretum: A Planner for Massive-Scale Federated Analytics with Differential Privacy

Elizabeth Margolin, Karan Newatia, Edo Roth, Tao Luo, and Andreas Haeberlen.
SOSP 2023, Koblenz, Germany
[PDF]

2021

Mycelium: Large-Scale Distributed Graph Queries with Differential Privacy

Edo Roth, Karan Newatia, Yiping Ma, Ke Zhong, Sebastian Angel, Andreas Haeberlen.
SOSP 2021, Virtual.
[PDF] [Talk] [Slides]

Do Not Overpay for Fault Tolerance!

Edo Roth, Andreas Haeberlen.
RTAS 2021, Nashville, TN, USA (Virtual).
[PDF] [Talk] [Slides]

REBOUND: Defending Distributed Systems Against Attacks with Bounded-Time Recovery

Neeraj Gandhi, Edo Roth,, Brian Sandler, Andreas Haeberlen, Linh Thi Xuan Phan.
EuroSys 2021, Edinburgh, UK (Virtual).
[PDF] [Talk] [Slides]

2020

Orchard: Differentially Private Analytics at Scale

Edo Roth, Hengchu Zhang, Andreas Haeberlen, Benjamin Pierce.
OSDI 2020, Banff, Canada (Virtual).
[PDF] [Talk] [Slides]

Testing Differential Privacy with Dual Interpreters

Hengchu Zhang, Edo Roth, Benjamin Pierce, Aaron Roth, Andreas Haeberlen.
OOPSLA 2020, Chicago, IL, USA.
[PDF]

Bounded-Time Recovery for Distributed Real-Time Systems

Neeraj Gandhi, Edo Roth, Robert Gifford, Linh Thi Xuan Phan, Andreas Haeberlen.
RTAS 2020, Sydney, Australia.
[PDF]

2019

Honeycrisp: Large-scale Differentially Private Aggregation Without a Trusted Core

Edo Roth, Daniel Noble, Andreas Haeberlen, Brett Hemenway.
SOSP 2019, Huntsville, Ontario, Canada.
[PDF] [Slides] [Talk]

Fuzzi: A Three-Level Logic for Differential Privacy

Hengchu Zhang, Edo Roth, Andreas Haeberlen, Benjamin C Pierce, Aaron Roth.
ICFP 2019, Berlin, Germany.
[PDF]

Teaching (as TA)

NETS 212: Scalable and Cloud Computing

Fall 2018, University of Pennsylvania, [Website]

CIS 511: Theory of Computation

Spring 2019, University of Pennsylvania

Service

Reviewer at AAAI Workshop on Privacy-Preserving Artificial Intelligence 2025

External Reviewer at PETS 2022, 2023

AEC Reviewer at EuroSys 2021

AEC Reviewer at OSDI 2020