I’m broadly interested in understanding how public data curators embed socially desirable values like privacy and confidentiality protections, equity, and reproducibility into their data publishing practices. My methodological research combines tools from theoretical computer science and computational social science to design and characterize complex structured errors induced by these practices. In doing so, I aim to demonstrate how these data curator interventions affect reproducible social science and evidence-based policymaking. Additionally, my qualitative research investigates the sociological and normative dimensions of how these interventions are implemented in practice; in particular, I’m interested in translational gaps between formal mathematical approaches and sociological approaches to ethics and values in data publishing, especially as applied to law and policy.
Selected recent works:
Seeman and Susser. "Between Privacy and Utility: On Differential Privacy in Theory and Practice." ACM Journal on Responsible Computing, 2023. [preprint]
Seeman, Sexton, Pujol, and Machanavajjhala. ``Privately Answering Queries on Skewed Data via Per Record Differential Privacy." Presented at Theory and Practice of Differential Privacy, under review at VLDB. [preprint][poster]
Seeman. ``Bettery Privacy Theorists for Better Data Stewards." Accepted at Journal of Privacy and Confidentiality, 2023. [preprint]
Seeman. "Private Treatment Assignment for Causal Experiments." Presented at Theory and Practice of Differential Privacy, under review at Journal of Privacy and Confidentiality, 2023. [abstract][poster]
Seeman. ``Framing Effects in the Operationalization of Differential Privacy Systems as Code-Driven Law." Proceedings of International Conference on Computer Ethics: Philosophical Enquiry, 2023. [article]
Seeman, Reimherr, and Slavkovic. "Formal Privacy for Partially Private Data." Under revision at Journal of Machine Learning Research, 2022. [preprint]
Seeman, Reimherr, and Slavkovic. "Exact Privacy Guarantees for Sampling Algorithms Implementing the Exponential Mechanism." Advances in Neural Information Processing Systems, 2021. [article]
Selected talks:
Seeman. "Interfacing Statistics and DP: Method and Mess." Keynote, International Conference on Theory and Practice of Differential Privacy (TPDP), Boston, MA, 2023. [slides]
Seeman. "Misspecification and Uncertainty Quantification in Differential Privacy." Invited Talk, Fields Institute Workshop on Differential Privacy and Statistical Data Analysis, Fields Institute, Toronto, ON, 2023. [video]
For a complete list, you can find my CV here and find my work on: