Hyperbolic Manifold Constrained Tabular Neural Network
Abstract
Tabular prediction is central to a wide range of real-world applications. Tabular data typically contain heterogeneous features as well as rich and complex relational information that can imply a latent structural manifold. Hyperbolic geometry can help capture complex structural relations in data. However, most existing tabular prediction models are constructed and optimized in Euclidean space. How to incorporate hyperbolic geometry into supervised tabular learning remains underexplored. We propose \textbf{HTNN}, a supervised hyperbolic manifold constrained tabular neural network for tabular prediction. HTNN consists of a hyperbolic feature-value representation layer for heterogeneous categorical and numerical features, followed by a conventional MLP predictor. HTNN employs a \emph{geometry-aware training} and \emph{geometry-free inference} optimization framework. The \emph{geometry-aware training} allows hyperbolic geometry to shape the latent representation learning of heterogeneous feature values. After training, the latent hyperbolic representations can be converted into ordinary Euclidean space for efficient \emph{geometry-free inference}. We conducted extensive experiments on the TALENT benchmark. HTNN ranks first among 36 methods on 200 classification datasets and third among 34 methods on 100 regression datasets. Experimental results show that the proposed hyperbolic manifold constrained tabular neural network is effective.
Related
- A diffusion-free multi-layer neural-network method for multidimensional nonlinear hyperbolic equations
- Kähler landscapes for complex neural network descents and guarantees including a search and destroy of the Calabi-Yau manifold
- Constrained defined-contribution pension management: A reparameterized neural network approach
- Soliton Profiles in 1D: Enhanced Neural-Network Solvers and Constrained Deep Ritz Methods
- The hyperbolic class and vanishing laws in a TMF-valued four-manifold invariant
BibTeX
@misc{li2026hyperbolic,
title = {{Hyperbolic Manifold Constrained Tabular Neural Network}},
author = {Tian Li and Huizhi Liang and Lucy Robinson and Varun Ojha},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2607.09710},
eprint = {2607.09710},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2607.09710},
}
RIS
TY - UNPB
TI - Hyperbolic Manifold Constrained Tabular Neural Network
AU - Tian Li
AU - Huizhi Liang
AU - Lucy Robinson
AU - Varun Ojha
PY - 2026
DA - 2026/09/28
PB - arXiv
DO - 10.48550/arxiv.2607.09710
UR - https://arxiv.org/abs/2607.09710
AB - Tabular prediction is central to a wide range of real-world applications. Tabular data typically contain heterogeneous features as well as rich and complex relational information that can imply a latent structural manifold. Hyperbolic geometry can help capture complex structural relations in data. However, most existing tabular prediction models are constructed and optimized in Euclidean space. How to incorporate hyperbolic geometry into supervised tabular learning remains underexplored. We propose \textbf{HTNN}, a supervised hyperbolic manifold constrained tabular neural network for tabular prediction. HTNN consists of a hyperbolic feature-value representation layer for heterogeneous categorical and numerical features, followed by a conventional MLP predictor. HTNN employs a \emph{geometry-aware training} and \emph{geometry-free inference} optimization framework. The \emph{geometry-aware training} allows hyperbolic geometry to shape the latent representation learning of heterogeneous feature values. After training, the latent hyperbolic representations can be converted into ordinary Euclidean space for efficient \emph{geometry-free inference}. We conducted extensive experiments on the TALENT benchmark. HTNN ranks first among 36 methods on 200 classification datasets and third among 34 methods on 100 regression datasets. Experimental results show that the proposed hyperbolic manifold constrained tabular neural network is effective.
ER -