A Sobering Look at Tabular Data Generation via Probabilistic Circuits
This work addresses the challenge of generating realistic tabular data for machine learning applications, revealing that current state-of-the-art methods may be overestimated due to inadequate metrics.
The paper questions the perception of progress in tabular data generation by highlighting limitations in current evaluation protocols and showing that a simple baseline, deep probabilistic circuits, achieves competitive or superior performance to state-of-the-art diffusion models at a fraction of the cost.
Tabular data is more challenging to generate than text and images, due to its heterogeneous features and much lower sample sizes. On this task, diffusion-based models are the current state-of-the-art (SotA) model class, achieving almost perfect performance on commonly used benchmarks. In this paper, we question the perception of progress for tabular data generation. First, we highlight the limitations of current protocols to evaluate the fidelity of generated data, and advocate for alternative ones. Next, we revisit a simple baseline -- hierarchical mixture models in the form of deep probabilistic circuits (PCs) -- which delivers competitive or superior performance to SotA models for a fraction of the cost. PCs are the generative counterpart of decision forests, and as such can natively handle heterogeneous data as well as deliver tractable probabilistic generation and inference. Finally, in a rigorous empirical analysis we show that the apparent saturation of progress for SotA models is largely due to the use of inadequate metrics. As such, we highlight that there is still much to be done to generate realistic tabular data. Code available at https://github.com/april-tools/tabpc.