Variational Deep Survival Machines: Survival Regression with Censored Outcomes

TL;DR

This paper introduces a novel variational auto-encoder based survival regression model that improves clustering of survival data and enhances long-term prediction accuracy, outperforming previous methods on benchmark datasets.

Contribution

The paper presents two variants of VAE for survival analysis, integrating clustering with survival prediction in an end-to-end trainable model, which is a novel approach.

Findings

Effective clustering of survival data achieved

Competitive scores on SUPPORT and FLCHAIN datasets

Superior long-term prediction performance

Abstract

Survival regression aims to predict the time when an event of interest will take place, typically a death or a failure. A fully parametric method [18] is proposed to estimate the survival function as a mixture of individual parametric distributions in the presence of censoring. In this paper, We present a novel method to predict the survival time by better clustering the survival data and combine primitive distributions. We propose two variants of variational auto-encoder (VAE), discrete and continuous, to generate the latent variables for clustering input covariates. The model is trained end to end by jointly optimizing the VAE loss and regression loss. Thorough experiments on dataset SUPPORT and FLCHAIN show that our method can effectively improve the clustering result and reach competitive scores with previous methods. We demonstrate the superior result of our model prediction in the long-term. Our code is available at https://github.com/qinzzz/auton-survival-785.