Deep End-to-End Survival Analysis with Temporal Consistency

TL;DR

This paper introduces a novel deep survival analysis method that leverages reinforcement learning principles and temporal consistency to model long-term dependencies in large-scale longitudinal data, improving stability and performance.

Contribution

The proposed framework uniquely integrates temporal consistency into deep survival analysis, enabling end-to-end training with complex architectures for better long-term pattern modeling.

Findings

Outperforms benchmarks on long sequence datasets

Enhances training stability through ablation studies

Effectively captures long-term temporal dependencies

Abstract

In this study, we present a novel Survival Analysis algorithm designed to efficiently handle large-scale longitudinal data. Our approach draws inspiration from Reinforcement Learning principles, particularly the Deep Q-Network paradigm, extending Temporal Learning concepts to Survival Regression. A central idea in our method is temporal consistency, a hypothesis that past and future outcomes in the data evolve smoothly over time. Our framework uniquely incorporates temporal consistency into large datasets by providing a stable training signal that captures long-term temporal relationships and ensures reliable updates. Additionally, the method supports arbitrarily complex architectures, enabling the modeling of intricate temporal dependencies, and allows for end-to-end training. Through numerous experiments we provide empirical evidence demonstrating our framework's ability to exploit temporal consistency across datasets of varying sizes. Moreover, our algorithm outperforms benchmarks on datasets with long sequences, demonstrating its ability to capture long-term patterns. Finally, ablation studies show how our method enhances training stability.