NeuralSurv: Deep Survival Analysis with Bayesian Uncertainty Quantification

TL;DR

NeuralSurv is a deep survival analysis model that incorporates Bayesian uncertainty quantification, improving calibration and providing robust uncertainty estimates in survival predictions.

Contribution

It introduces a novel non-parametric, architecture-agnostic framework with a two-stage data-augmentation scheme and a scalable variational inference algorithm for Bayesian deep survival analysis.

Findings

NeuralSurv achieves superior calibration compared to existing models.

It matches or exceeds the discriminative performance of state-of-the-art models.

The model provides well-calibrated uncertainty estimates, especially in data-scarce regimes.

Abstract

We introduce NeuralSurv, the first deep survival model to incorporate Bayesian uncertainty quantification. Our non-parametric, architecture-agnostic framework captures time-varying covariate-risk relationships in continuous time via a novel two-stage data-augmentation scheme, for which we establish theoretical guarantees. For efficient posterior inference, we introduce a mean-field variational algorithm with coordinate-ascent updates that scale linearly in model size. By locally linearizing the Bayesian neural network, we obtain full conjugacy and derive all coordinate updates in closed form. In experiments, NeuralSurv delivers superior calibration compared to state-of-the-art deep survival models, while matching or exceeding their discriminative performance across both synthetic benchmarks and real-world datasets. Our results demonstrate the value of Bayesian principles in data-scarce regimes by enhancing model calibration and providing robust, well-calibrated uncertainty estimates for the survival function.