Adaptive Transformer Modelling of Density Function for Nonparametric Survival Analysis

TL;DR

This paper introduces UniSurv, a Transformer-based survival analysis model that generates high-quality unimodal probability density functions without prior assumptions, effectively handling both static and dynamic data.

Contribution

The paper presents a novel Transformer-based survival regression method, UniSurv, which models unimodal PDFs and improves sensitivity in censoring prediction without relying on prior distribution assumptions.

Findings

UniSurv outperforms existing methods in censoring prediction.

It effectively models both temporal and non-temporal data.

Demonstrates higher emphasis on censoring in experiments.

Abstract

Survival analysis holds a crucial role across diverse disciplines, such as economics, engineering and healthcare. It empowers researchers to analyze both time-invariant and time-varying data, encompassing phenomena like customer churn, material degradation and various medical outcomes. Given the complexity and heterogeneity of such data, recent endeavors have demonstrated successful integration of deep learning methodologies to address limitations in conventional statistical approaches. However, current methods typically involve cluttered probability distribution function (PDF), have lower sensitivity in censoring prediction, only model static datasets, or only rely on recurrent neural networks for dynamic modelling. In this paper, we propose a novel survival regression method capable of producing high-quality unimodal PDFs without any prior distribution assumption, by optimizing novel Margin-Mean-Variance loss and leveraging the flexibility of Transformer to handle both temporal and non-temporal data, coined UniSurv. Extensive experiments on several datasets demonstrate that UniSurv places a significantly higher emphasis on censoring compared to other methods.