Olive Franzese-McLaughlin

About Me
Research
My work is about auditing AI/ML systems without leaking sensitive
information, by using secure cryptographic protocols. This allows
auditors to provably verify that AI/ML systems are reliable, unbiased,
and privacy-preserving, even in the presence of
misaligned
incentives
and/or
adversarial
behavior.
It also allows service providers build trust by actively
attesting that their models are trustworthy, without revealing any
information about user data or proprietary model parameters. I believe
techniques like these are integral building blocks for effective
regulation of AI/ML.
My published works in this area include methods for cryptographic
verification of fairness
(Alan
Turing Institute Research Highlight, + followup
work), confidence calibration, differential privacy, and robustness using zero-knowledge proofs and
secure multiparty computation. I’ve worked on efficient zero-knowledge
proof building blocks as well.
I have also contributed work on data-driven computational biology
algorithms for cancer diagnostics that are more
accessible to patients, signaling pathway
extraction from protein-protein interaction data, and systematic experiment planning. In ancient
history, I was a wet lab biologist.
Publications
- Secure Noise Sampling for
Differentially Private Collaborative
Learning Olive
Franzese, C Fang, R Garg, S Jha, N Papernot, X Wang, A
Dziedzic. Accepted to ACM CCS ’25.
- Confidential Guardian:
Cryptographically Prohibiting the Abuse of Model
Abstention S Rabanser, AS
Shamsabadi, Olive Franzese, X Wang, A Weller, N
Papernot. Accepted to ICML ’25.
- OATH: Efficient and
Flexible Zero-Knowledge Proofs of End-to-End ML
Fairness Olive Franzese, AS
Shamsabadi, H Haddadi. arXiv preprint.
- Robust and actively secure
serverless collaborative learning
Olive Franzese, A Dziedzic, C Choquette-Choo, M Thomas,
M Kaleem, S Rabanser, C Fang, S Jha, N Papernot, X Wang. NeurIPS 2023.
- Confidential-PROFITT:
Confidential PROof of FaIr Training of Trees
AS Shamsabadi, S Wyllie, Olive Franzese, N
Dullerud, S Gambs, N Papernot, X Wang, A Weller. ICLR 2023. Notable
Top 5% Commendation.
Alan Turing Institute Top 10 Research Highlight of
2022-23.
- Constant-overhead zero-knowledge for
RAM programs Olive Franzese, J
Katz, S Lu, R Ostrovsky, X Wang, C Weng. ACM CCS 2021.
- ScalpelSig Designs Targeted
Genomic Panels from Data to Detect Activity of Mutational
Signatures Olive
Franzese, J Fan, R Sharan, MDM Leiserson. RECOMB 2021, Journal
of Computational Biology 2022.
- Hypergraph-based
connectivity measures for signaling pathway
topologies Olive
Franzese, A Groce, TM Murali, A Ritz. PLoS Comp Biol 2019.
- CrossPlan: systematic
planning of genetic crosses to validate mathematical
models A Pratapa, N Adames, P Kraikivski,
Olive Franzese, J J Tyson, J Peccoud, TM Murali.
Bioinformatics 2018.
- Stem
Cell-Like Dog Placenta Cells Afford Neuroprotection against Ischemic
Stroke Model via Heat Shock Protein Upregulation
S Yu, N Tajiri, Olive Franzese, M Franzblau, E Bae, S
Platt, Y Kaneko, C V Borlongan. PLoS One 2013.
- Human
umbilical cord blood for transplantation therapy in myocardial
infarction S A Acosta, Olive
Franzese, M Staples, N L Weinbren, M Babilonia, J Patel, N
Merchant, A J Simancas, A Slakter, M Caputo, M Patel, G Franyuti, M H
Franzblau, L Suarez, C Gonzales-Portillo, T Diamandis, K Shinozuka, N
Tajiri, P R Sanberg, Y Kaneko, L W Miller, C V Borlongan. Journal of
stem cell research & therapy 2013.
- Stroke
in the eye of the beholder H Ishikawa, M Caputo,
Olive Franzese, N L Weinbren, A Slakter, M Patel, C E
Stahl, M A Jacotte, S Acosta, G Franyuti, K Shinozuka, N Tajiri, H van
Loveren, Y Kaneko, C V Borlongan. Med Hypotheses 2013.
- Nestin
overexpression precedes caspase-3 upregulation in rats exposed to
controlled cortical impact traumatic brain injury
Y Kaneko, N Tajiri, S Yu, T Hayashi, C E Stahl, E Bae, H Mestre,
Olive Franzese, A Rodrigues Jr, M C Rodrigues, H
Ishikawa, K Shinozuka, W Hethorn, N Weinbren, L E Glover, J Tan, A H
Achyuta, H Van Loveren, P R Sanberg, S Shivsankar, C V Borlongan. Cell
medicine 2012.
Funding
I am a
Vector
Distinguished Postdoctoral Fellow, an
NSF GRFP fellow, and a
National
Cancer Institute CRTA recipient.
Teaching & Mentorship
Teaching and mentoring students is very important to me. In my
pedagogy, I aim to expose the deep and compelling questions that
underlie computing, rather than just teaching about the specifics of
computing systems themselves. I take a lot of inspiration from the
writings of contemporary mathematics educator
Paul
Lockhart, and seek to apply his insights in computer science
education. I am especially interested in the developing role of
computing as part of a liberal arts education. I think we are in great
need of computing practitioners who can reason about the social context
surrounding computing, in addition to programs themselves. I think a
liberal arts education is exactly the right place to cultivate students
with mastery of both. I was on the ACM SIGSCE 2025
Committee on
Computing Education in Liberal Arts Colleges. Below are some of my
experiences in undergraduate teaching and mentorship.
Research Advisees
- Patrick Norton (Incoming Post-Bac Internship 2025)
- Carter Luck (Post-Bac Internship 2025 → PhD UMass Amherst)
- Elle Wen (Undergraduate Thesis 2023-24 → MSci Waterloo) - Fuzzy
private set intersection for satellite collision detection
- Lydia Longfritz (Undergraduate Internship 2023) - Secure multiparty
computation for privacy-preserving mutational signature analysis
- John Poole (Undergraduate Thesis 2022-23 → MPhil Cambridge) -
Cellular automata simulation of cancer ecology
Visiting Professor, Reed
College
- CSCI 221: COMPUTER SCIENCE FUNDAMENTALS II
- an introductory systems-level programming course in C, MIPS
Assembly, and C++.
- selected student comments:
“I loved the energy of the lectures and the learning environment,
the professor’s enthusiasm in providing help in lab, and the
effectiveness of the assignment material at both closely matching the
subjects discussed in lecture while providing sufficient practice to
reinforce learning. The midterm exam was also extremely well-designed
and well- structured - in fact, it felt like one of the most accurately
balanced exams I have taken at Reed. I would retain the majority of the
course structure and policies - I feel that the course accomplished
exactly what it is meant to, with the appropriate scope at all steps of
the process, and with effective measures in place to ensure students
have a good experience.”
“I liked Olive’s teaching style! She was always super prepared,
organized, cheerful, and engaged during lectures and labs. The policy of
going and offering help unsolicited during labs was very helpful for
folks such as I who struggle asking for help for any reason ever. Asking
questions unsolicited during lecture helped for this exact same reason,
it never felt like being singled out or stressful as the answer “I don’t
know” was never punished. It was clear that she cares, not just whether
the students learned the material, but also whether we [were] generally
mentally well. I felt respected and appreciated every day I showed up to
class, and though I’ve stopped going out of my way to give a 110 percent
in college, this class was the one exception.”
- full course
evaluation here
Graduate TA,
Northwestern & University of Maryland
- CS 307: INTRODUCTION TO CRYPTOGRAPHY (2021)
- CMSC 132: OBJECT-ORIENTED PROGRAMMING II
(2019)
- CMSC 131: OBJECT-ORIENTED PROGRAMMING I (2018)
Undergraduate
Teaching Assistant, Reed College
- MATH 121: INTRODUCTION TO COMPUTING
(2017-2019)