Differentially Private (tabular) Generative Models Papers with Code

Copulas

• DPCopula | Differentially Private Synthesization of Multi-Dimensional Data using Copula Functions (Li et al., 2014) | code
• dpc | Differentially Private Release of High-Dimensional Datasets using the Gaussian Copula (Asghar et al., 2020)
• Copula-Shirley | Growing Synthetic Data Through Differentially-Private Vine Copulas (Gambs et al., 2021) | code | video

Marginals

• DPSyn | DPSyn: Experiences in the NIST Differential Privacy Data Synthesis Challenges (Li et al., 2021) | code code_1
• PrivSyn | PrivSyn: Differentially Private Data Synthesis (Zhang et al., 2021)

Graphical Models

• PrivBayes | PrivBayes: Private Data Release via Bayesian Networks (Zhang et al., 2017) | code (C++), reimplementation (Python)
• PrivMN | Differentially Private High-Dimensional Data Publication via Markov Network (Zhang et al., 2019)
• MST | Winning the NIST Contest: A Scalable and General Approach to Differentially Private Synthetic Data (McKenna et al., 2021) | code
• PrivMRF | Data Synthesis via Differentially Private Markov Random Fields (Cai et al., 2021) | code
• dpart | dpart: Differentially Private Autoregressive Tabular, a General Framework for Synthetic Data Generation (Mahiou et al., 2022) | code

VAEs

• DPGM | Differentially Private Mixture of Generative Neural Networks (Acs et al., 2017)
• DP-SYN | Privacy Preserving Synthetic Data Release Using Deep Learning (Abay et al., 2018)
• P3GM | P3GM: Private High-Dimensional Data Release via Privacy Preserving Phased Generative Model (Takagi et al., 2021) | code

GANs

• DPGAN | Differentially Private Generative Adversarial Network (Xie et al., 2018) | code
• PATE-GAN | PATE-GAN: Generating Synthetic Data with Differential Privacy Guarantees (Jordon et al., 2019) | code, code_1
• dp-GAN-TSCD | Differentially Private Generative Adversarial Networks for Time Series, Continuous, and Discrete Open Data (Frigerio et al., 2019) | code
• SPRINT-gan | Privacy-Preserving Generative Deep Neural Networks Support Clinical Data Sharing (Beaulieu-Jones et al., 2019) | code
• DPautoGAN | Differentially Private Synthetic Mixed-Type Data Generation For Unsupervised Learning (Tantipongpipat et al., 2021) | code
• G-PATE | G-PATE: Scalable Differentially Private Data Generator via Private Aggregation of Teacher Discriminators (Long et al., 2021) | code

Transformers

• DP-TBART | DP-TBART: A Transformer-based Autoregressive Model for Differentially Private Tabular Data Generation (Castellon et al., 2023)

Other Approaches

• RON-Gauss | RON-Gauss: Enhancing Utility in Non-Interactive Private Data Release (Chanyaswad et al., 2019) | code
• FEM | New Oracle-Efficient Algorithms for Private Synthetic Data Release (Vietri et al., 2020) | code
• RAP | Differentially Private Query Release Through Adaptive Projection (Aydore et al., 2021) | code
• Kamino | Kamino: Constraint-Aware Differentially Private Data Synthesis (Ge et al., 2021) | code
• PEP, GEM | Iterative Methods for Private Synthetic Data: Unifying Framework and New Methods (Liu et al., 2021) | code, code_1
• AIM | AIM: An Adaptive and Iterative Mechanism for Differentially Private Synthetic Data (McKenna et al., 2022) | code
• RAP++ | Private Synthetic Data for Multitask Learning and Marginal Queries (Vietri et al., 2022)
• NAPSU-MQ | Noise-Aware Statistical Inference with Differentially Private Synthetic Data (Räisä et al., 2023) | code
• Private-GSD | Generating Private Synthetic Data with Genetic Algorithms (Liu et al., 2023) | code
• CuTS | CuTS: Customizable Tabular Synthetic Data Generation (Vero et al., 2024) | code

Other NIST Differential Privacy Challenges Solutions

• DP-WGAN | Differentially Private Dataset Release using Wasserstein GANs (Alzantot et al., 2019) | code
• DPFieldGroups | Differential Privacy Synthetic Data Challenge Algorithm (Gardner, 2019) | code

Other Data Domains

• GANobfuscator | GANobfuscator: Mitigating Information Leakage Under GAN via Differential Privacy (Xu et al., 2019)
• DP-CGAN | DP-CGAN: Differentially Private Synthetic Data and Label Generation (Torkzadehmahani et al., 2019) | code
• GS-WGAN | GS-WGAN: A Gradient-Sanitized Approach for Learning Differentially Private Generators (Chen et al., 2020) | code
• DPDoppelGANger | Using GANs for Sharing Networked Time Series Data: Challenges, Initial Promise, and Open Questions (Lin et al., 2020) | code
• DP-Sinkhorn | Don't Generate Me: Training Differentially Private Generative Models with Sinkhorn Divergence (Cao et al., 2021)
• DPDMs | Differentially Private Diffusion Models (Dockhorn et al., 2023) | code
• DP-LDMs | Differentially Private Latent Diffusion Models (Lyu et al., 2023) | code

Non Peer-Reviewed Models

• dp-GAN | Differentially Private Releasing via Deep Generative Model (Technical Report) (Zhang et al., 2018) | code
• DP-AuGM, DP-VaeGM | Differentially Private Data Generative Models (Chen et al., 2018)
• Differentially Private Diffusion Models Generate Useful Synthetic Images (Ghalebikesabi et al., 2023)
• PPSyn | Partition-based differentially private synthetic data generation (Zhang et al., 2023)
• TableDiffusion | Generating tabular datasets under differential privacy (Truda, 2024) | code
• DP-LLMTGen | Differentially Private Tabular Data Synthesis using Large Language Models (Tran & Xiong, 2024)