Summary of Interpretable Machine Learning For Survival Analysis, by Sophie Hanna Langbein et al.
Interpretable Machine Learning for Survival Analysis
by Sophie Hanna Langbein, Mateusz Krzyziński, Mikołaj Spytek, Hubert Baniecki, Przemysław Biecek, Marvin N. Wright
First submitted to arxiv on: 15 Mar 2024
Categories
- Main: Machine Learning (stat.ML)
- Secondary: Machine Learning (cs.LG); Methodology (stat.ME)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed paper presents a comprehensive review of interpretable machine learning (IML) methods for survival analysis, with a focus on black box models. The authors highlight the importance of IML in medical and healthcare contexts, where transparency and accountability are crucial. They demonstrate how existing IML methods, such as individual conditional expectation (ICE), partial dependence plots (PDP), accumulated local effects (ALE), and feature importance measures can be adapted to survival outcomes. The paper also showcases an application of these methods to real-world data on under-5 year mortality in Ghana. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This research reviews ways to make machine learning models more understandable, especially when predicting when events will happen. Doctors, policymakers, and researchers need this kind of transparency because it helps them identify potential biases or limitations in the models. The authors explore different methods that can be used to explain survival analysis, including individual conditional expectation, partial dependence plots, and accumulated local effects. They also apply these methods to real data on child mortality in Ghana. |
Keywords
* Artificial intelligence * Machine learning
