Hi, I'm Shlomi.

I’m a CS Ph.D. candidate at Boston University, under the supervision of Prof. Ran Canetti. Since September 2023, I have been visiting Columbia University to collaborate with Prof. Rachel Cummings.

I’m working on Responsible AI, and my current interests include:

1. Designing differentially private synthetic data for use in government and medical contexts
2. The co-design of computer science and law, focusing on privacy and fairness
3. Supporting policymakers in developing effective AI policies

In February 2024, together with the Israeli Ministry of Health, we released a differentially private synthetic dataset of the National Live Birth Registry.

For the last few years, I have taught courses in Responsible AI, Law, Ethics & Society in various institutes including Boston University, Cornell Tech, Bocconi University, Tel Aviv University and the Technion. Our materials are available for faculty here. In August 2023, I taught a two-day congressional workshop for US Congress staffers based on our course.

In summer 2022, I interned at Twitter Cortex where I leveraged human-in-the-loop research to improve toxicity models. In 2020-2021, I was an Associated Researcher at the Alexander von Humboldt Institute for Internet and Society (HIIG) in Berlin. In summer 2019, I did a research internship at the Center for Human-Compatible AI at UC Berkeley, working on neural network interpretability.

In my previous life, I was a social entrepreneur - co-founder of the Israeli Cyber Education Center. There I led the development of nationwide educational programs in computing for kids and teens. The center aims to increase the social mobility of underrepresented groups in tech, such as women, minorities, and individuals from the suburbs of Israel. I co-authored a Computer Network textbook in a tutorial approach (in Hebrew). Before that, I was an algorithmic research team leader in cybersecurity.

Publications

Rachel Cummings, SH, Jayshree Sarathy, Marika Swanberg. “Attaxonomy: Unpacking Differential Privacy Guarantees Against Practical Adversaries”. To appear in the non-archival track of the Symposium on Foundations of Responsible Computing - FORC (2024).

SH, Karni Chagal-Feferkorn, Niva Elkin-Koren and Avigdor Gal. “Data Science Meets Law: Learning Responsible AI Together”. Communications of the ACM (2022). Featured on the journal cover.

*Gavin Brown, *SH, *Iden Kalemaj. “Performative Prediction in a Stateful World”. International Conference on Artificial Intelligence and Statistics - AISTATS (2022). Preliminary version at NeurIPS Workshop on Consequential Decision Making in Dynamic Environments, with contributed talk (2020).

*Stephen Casper, *SH, *Daniel Filan, Cody Wild, Andrew Critch and Stuart Russell. “Graphical Clusterability and Local Specialization in Deep Neural Networks”. ICLR PAIR²Struct Workshop (2022).

*SH, *Stephen Casper, *Daniel Filan, Cody Wild, Andrew Critch and Stuart Russell. “Detecting Modularity in Deep Neural Networks”. arXiv preprint arXiv:2110.08058 (2021).

*Daniel Filan, *Stephen Casper, *SH, Cody Wild, Andrew Critch, and Stuart Russell. “Clusterability in Neural Networks”. arXiv preprint arXiv:2103.03386 (2021).