[CIKM '24] SurvReLU: Inherently Interpretable Survival Analysis via Deep ReLU Networks

News 🔥

• July 16, 2024: Congratulations! This paper has been accepted by CIKM 2024 $\color{lightgreen} \textbf{[acceptance rate = 27\%]}%$ !

Abstract. Survival analysis models time-to-event distributions with censorship. Recently, deep survival models using neural networks have dominated due to their representational power and state-of-the-art performance. However, their "black-box" nature hinders interpretability, which is crucial in real-world applications. In contrast, "white-box" tree-based survival models offer better interpretability but struggle to converge to global optima due to greedy expansion. In this paper, we bridge the gap between previous deep survival models and traditional tree-based survival models through deep rectified linear unit (ReLU) networks. We show that a deliberately constructed deep ReLU network (SurvReLU) can harness the interpretability of tree-based structures with the representational power of deep survival models. Empirical studies on both simulated and real survival benchmark datasets show the effectiveness of the proposed SurvReLU in terms of performance and interoperability.

Network Architecture

Usage

1. Prepare dataset

2. Dependencies

3. Train From Scratch

Acknowledgements

This code is built on the top of pycox and locally constant network, we thank their efficient and neat codebase.

Citation

If you find our work is useful in your research, please consider raising a star ⭐ and citing:

@misc{sun2024survreluinherentlyinterpretablesurvival,
      title={SurvReLU: Inherently Interpretable Survival Analysis via Deep ReLU Networks}, 
      author={Xiaotong Sun and Peijie Qiu and Shengfan Zhang},
      year={2024},
      eprint={2407.14463},
      archivePrefix={arXiv},
      primaryClass={cs.LG},
      url={https://arxiv.org/abs/2407.14463}, 
}