Visit my Google Scholar page for the full list of my publications.

Publications

Georgi Ganev, Emiliano De Cristofaro
The Inadequacy of Similarity-based Privacy Metrics: Privacy Attacks against “Truly Anonymous” Synthetic Datasets
46th IEEE Symposium on Security & Privacy (IEEE S&P 2025)
Distinguished Paper
pre-print | paper
Coverage: FCA, International AI Safety Report, Ted’s blog, HackerNoon

Georgi Ganev, Meenatchi Sundaram Muthu Selva Annamalai, Emiliano De Cristofaro
The Elusive Pursuit of Reproducing PATE-GAN: Benchmarking, Auditing, Debugging
Transactions on Machine Learning Research (TMLR 2025)
pre-print | paper | code

Meenatchi Sundaram Muthu Selva Annamalai, Georgi Ganev, Emiliano De Cristofaro
“What do you want from theory alone?” Experimenting with Tight Auditing of Differentially Private Synthetic Data Generation
33rd USENIX Security Symposium (USENIX Security 2024)
pre-print | paper | video | code

Georgi Ganev, Kai Xu, Emiliano De Cristofaro
Graphical vs. Deep Generative Models: Measuring the Impact of Differentially Private Mechanisms and Budgets on Utility
31st ACM Conference on Computer and Communications Security (ACM CCS 2024)
pre-print | paper

Kai Xu, Georgi Ganev, Emile Joubert, Rees Davison, Olivier Van Acker, Luke Robinson
Synthetic Data Generation of Many-to-Many Datasets via Random Graph Generation
11th International Conference on Learning Representations (ICLR 2023)
pre-print | paper

Georgi Ganev, Bristena Oprisanu, Emiliano De Cristofaro
Robin Hood and Matthew Effects: Differential Privacy Has Disparate Impact on Synthetic Data
39th International Conference on Machine Learning (ICML 2022)
pre-print | paper | poster | video
Coverage: Apple, Nokia Bell Labs

Bristena Oprisanu, Georgi Ganev, Emiliano De Cristofaro
On Utility and Privacy in Synthetic Genomic Data
(earlier version accepted at 8th Workshop on Genome Privacy and Security (GenoPri 2021))
29th Network and Distributed System Security Symposium (NDSS 2022)
pre-print | paper

Pre-prints

Georgi Ganev, Meenatchi Sundaram Muthu Selva Annamalai, Sofiane Mahiou, Emiliano De Cristofaro
The Importance of Being Discrete: Measuring the Impact of Discretization in End-to-End Differentially Private Synthetic Data
pre-print

Zexi Yao, Natasa Krco, Georgi Ganev, Yves-Alexandre de Montjoye
The DCR Delusion: Measuring the Privacy Risk of Synthetic Data
pre-print

Workshops

Georgi Ganev, Meenatchi Sundaram Muthu Selva Annamalai, Sofiane Mahiou, Emiliano De Cristofaro
Understanding the Impact of Data Domain Extraction on Synthetic Data Privacy
1st ICLR Synthetic Data x Data Access Problem workshop (SynthData 2025), part of ICLR 2025
pre-print | paper | poster

Georgi Ganev
Synthetic Data, Similarity-based Privacy Metrics, and Regulatory (Non-)Compliance
2nd ICML Workshop on Generative AI and Law (GenLaw 2024), part of ICML 2024
pre-print | paper | poster

Georgi Ganev
When Synthetic Data Met Regulation
1st ICML Workshop on Generative AI and Law (GenLaw 2023), part of ICML 2023
pre-print | paper | poster

Lauren Arthur, Jason Costello, Jonathan Hardy, Will O’Brien, James Rea, Gareth Rees, Georgi Ganev
On the Challenges of Deploying Privacy-Preserving Synthetic Data in the Enterprise
1st ICML Workshop on Deployment Challenges for Generative AI, part of ICML 2023
pre-print | paper
Coverage: Montreal AI Ethics Institute

Sofiane Mahiou, Kai Xu, Georgi Ganev
dpart: Differentially Private Autoregressive Tabular, a General Framework for Synthetic Data Generation
8th ICML Workshop on Theory and Practice of Differential Privacy (TPDP 2022), part of ICML 2022
pre-print | code | poster | video

Georgi Ganev
DP-SGD vs PATE: Which Has Less Disparate Impact on GANs?
3rd AAAI Workshop on Privacy-Preserving Artificial Intelligence (PPAI 2022), part of AAAI 2022
pre-print | paper | poster | video