The Synthetic Data Vault
By Neha Patki, Roy Wedge, Kalyan Veeramachaneni. 2016
748 citations recorded by the source corpus.
Subfields: Computer Networks and Communications, Management Science and Operations Research, Signal Processing
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computer Networks and Communications
- Samuel A. Assefa
- Yuji Roh
- Lei Xu
- Vadim Borisov
- Gjergji Kasneci
- Tobias Leemann
- Kathrin Sessler
- Martin Pawelczyk
In Management Science and Operations Research
- Samuel A. Assefa
- Lei Xu
- Vadim Borisov
- Gjergji Kasneci
- Tobias Leemann
- Kathrin Sessler
- Martin Pawelczyk
- Kalyan Veeramachaneni
In Signal Processing
- Samuel A. Assefa
- Yuji Roh
- Lei Xu
- Vadim Borisov
- Gjergji Kasneci
- Tobias Leemann
- Kathrin Sessler
- Martin Pawelczyk
Citations in the corpus
- Synthesizing Tabular Data using Generative Adversarial Networks (2018, methodology): “)P(y) , (6) where E() computes the entropy, and X,Yare two selected columns. Other data synthesizers: We compare our TGAN synthesizer with three published methods for generating synthetic data. 5 GC [34] uses a Gaussian Copula to hierarchically generate multiple synthetic tables in a database. BN-Id”
- A Survey on Data Collection for Machine Learning: A Big Data - AI Integration Perspective (2018, background): “Generating synthetic data along with labels is increasingly being used in machine learning due to its low cost and flexibility [90].”
- Modeling Tabular data using Conditional GAN (2019, methodology): “A set of non-linearly correlated continuous variables could be modeled using copulas [19, 23].”
- Generating synthetic data in finance: opportunities, challenges and pitfalls (2020, methodology): “The synthetic data vault [59] is a more contemporary approach, with a synthetic database specified via JSON.”
- Language Models are Realistic Tabular Data Generators (2022, methodology): “The CopulaGAN model from the Synthetic Data Vault (SDV) framework (Patki et al., 2016) is applying the Gaussian copulas to simplify the underlying CTGAN.”
- Generating Synthetic but Plausible Healthcare Record Datasets (2018, neutral): “See [9], for example, for compelling motivation. traditional way of generating synthetic datasets is to create a generative model from scratch using human knowledge, for example a simulator, then run it to generate as many instances as desired.”
- Deep Neural Networks and Tabular Data: A Survey (2021, methodology): “Classical approaches for the data generation task include Copulas [126], [127] and Bayesian networks [128].”