profile image

Max Cairney-Leeming

I am a PhD student at the Institute of Science and Technology Austria (ISTA), near Vienna, supervised by Christoph Lampert. I work on differential privacy for deep learning, and in particular on how privacy interacts with other trustworthiness concerns, such as fairness and worst-group performance. I also work on the theory of DP: what it guarantees against data reconstruction, and how it implies generalization. I am a member of the ELLIS PhD Program, co-supervised by Amartya Sanyal at the University of Copenhagen.

Preprint available: Data Reconstruction under DP

Our work with Simone Bombari and Marco Mondelli, “A Sharp Transition in Data Reconstruction under Differential Privacy”, is now available at arXiv:2609.37344.

September 2026

Posters at TPDP and FORC

I presented posters on private worst-group optimization (arXiv:2602.10820) at the Theory and Practice of Differential Privacy (TPDP) workshop and at the Symposium on Foundations of Responsible Computing (FORC), both in Boston.

June 2026

Preprint available: From Privacy to Generalization

Our work with Christoph Lampert and Hossein Zakerinia, “From Privacy to Generalization: Linear Max-Information Bounds for Differentially Private Learning Algorithms”, is now available at arXiv:2605.26222.

May 2026

Preprint available: Private Worst-Group Optimization

Our recent work on “Adaptive Sampling and Clipping for Private Worst-Case Group Optimization” is now available at arXiv:2602.10820. This work was jointly supervised by Amartya Sanyal and Christoph Lampert, and began during my visit to Copenhagen.

February 2026

PPML Workshop @ EurIPS

I presented a poster on a preliminary version of my work on private worst-group optimization (now available at arXiv:2602.10820) at the PPML workshop at EurIPS in Copenhagen, organized by several members of my group and close collaborators.

December 2025

Paper in TMLR: Demystifying Amortized Causal Discovery

Our paper “Demystifying amortized causal discovery with transformers”, with Francesco Montagna, Dhanya Sridhar and Francesco Locatello, has been published in Transactions on Machine Learning Research (OpenReview).

December 2025

Visit to the University of Copenhagen

I am visiting Amartya Sanyal’s group at the Datalogisk Institute at UCPH for the next 3 months, as part of my ELLIS exchange.

April 2025

ELSA Workshop in Bertinoro

I attended the ELSA Workshop on Privacy-Preserving Machine Learning (17–21 March) at the Bertinoro University Centre in Italy, for ELSA and ELLIS members.

March 2025

Qualifying Exam

I presented my PhD project proposal to my committee: Christoph Lampert, Marco Mondelli, Amartya Sanyal and Florian Praetorius (chair).

February 2025

Affiliated with the Lampert group

I have officially joined the group of Prof. Christoph Lampert at ISTA.

June 2024

Starting my PhD

I have started my PhD at ISTA! I’m looking forward to working with 3 different professors over the next year, to broaden my ML horizons, and conversely, to home in on a topic and affiliate with one of them for the rest of my time at ISTA.

September 2023

Graduation

I graduated from Oxford! I received a First (years 1-3; Bachelor’s) and a Distinction (4th year; Master’s) in my integrated MMathCompSci degree in Mathematics and Computer Science.