TabSurv: Adapting Modern Tabular Neural Networks to Survival Analysis

2026-05-05

AI Summary

This work addresses the limited transferability and task-specific nature of existing deep learning approaches in survival analysis. To overcome these constraints, the authors propose a general-purpose adaptation framework that integrates modern tabular neural networks into survival modeling, supporting both Weibull parametric and non-parametric prediction paradigms. A novel histogram-based loss function, SurvHL, is specifically designed to handle censored data. The study further introduces, for the first time in this domain, a deep ensemble of parallel-trained MLPs, optimizing distribution parameters before averaging to enhance predictive diversity. Evaluated across ten real-world datasets, the proposed method consistently outperforms established baselines—including Random Survival Forests (RSF), DeepSurv, DeepHit, and SurvTRACE—with the Weibull-based deep ensemble achieving the highest average C-index performance.

Abstract

Survival analysis on tabular data is a well-studied problem. However, existing deep learning methods are often highly task-specific, which can limit the transfer of new approaches from other domains and introduce constraints that may affect performance. We propose TabSurv, an approach that adapts modern tabular architectures to survival analysis using either the Weibull distribution or non-parametric survival prediction. TabSurv optimizes SurvHL, a novel histogram loss function supporting censored data. In addition to a baseline feed-forward network, we implement deep ensembles of MLPs for survival analysis within TabSurv. In contrast to prior work, the ensemble components are trained in parallel, optimizing survival distribution parameters before averaging, which promotes diversity across ensemble component predictions. We perform a comprehensive empirical evaluation of different proposed architectures on 10 diverse real-world survival datasets. Our results show that TabSurv consistently outperforms on average established classical and deep learning baselines, such as RSF, DeepSurv, DeepHit, SurvTRACE. Notably, deep ensembles with Weibull parametrization instead of non-parametric models achieve the highest average rank by C-index. Overall, our study clarifies how modern tabular neural networks can be adapted and trained to tackle survival analysis problems, offering a strong and reliable approach. The TabSurv implementation is publicly available.

Problem

• survival analysis
• tabular data
• deep learning
• neural networks
• censored data

Innovation

• TabSurv
• Survival Analysis
• Tabular Neural Networks
• Deep Ensembles
• Histogram Loss