TabDDPM: Modelling Tabular Data with Diffusion Models

Akim Kotelnikov, Dmitry Baranchuk, Ivan Rubachev, Artem Babenko

Keywords: tabular data, diffusion models, generative modelling

Abstract: Denoising diffusion probabilistic models are currently becoming the leading paradigm of generative modeling for many important data modalities. Being the most prevalent in the computer vision community, diffusion models have also recently gained some attention for other domains, including speech, NLP, and graph-like data. In this work, we investigate if the framework of diffusion models can be advantageous for general tabular problems, where datapoints are typically represented by vectors of heterogeneous features. The inherent heterogeneity of tabular data makes it quite challenging for accurate modeling, since the individual features can be of completely different nature, i.e., some of them can be continuous and some of them can be discrete. To address such data types, we introduce TabDDPM --- a diffusion model that can be universally applied to any tabular dataset and handles any types of features. We extensively evaluate TabDDPM on a wide set of benchmarks and demonstrate its superiority over existing GAN/VAE alternatives, which is consistent with the advantage of diffusion models in other fields. Additionally, we show that TabDDPM can be successfully used in privacy-oriented setups, where the original datapoints cannot be shared.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Submission Guidelines: Yes

Please Choose The Closest Area That Your Submission Falls Into: Generative models

TL;DR: Proposed a new state-of-the-art approach for tabular data generation using diffusion models

Community Implementations: [ 4 code implementations](https://www.catalyzex.com/paper/arxiv:2209.15421/code)