Summary of Conformalized Survival Distributions: a Generic Post-process to Increase Calibration, by Shi-ang Qi et al.
Conformalized Survival Distributions: A Generic Post-Process to Increase Calibration
by Shi-ang Qi, Yakun Yu, Russell Greiner
First submitted to arxiv on: 12 May 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
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 This paper proposes a novel approach to survival analysis that optimizes both discrimination and calibration simultaneously. The authors argue that previous models often trade-off between these two properties, improving one while degrading the other. To address this issue, they introduce conformal regression, which can improve a model’s calibration without sacrificing its ability to accurately rank subjects. The authors provide theoretical guarantees for their approach and validate its effectiveness across 11 real-world datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps survival models do two important things: correctly rank people who will survive longer (discrimination) and make accurate predictions about when events will happen (calibration). Usually, these two tasks are hard to balance. The researchers came up with a new way called conformal regression that does better on both tasks at the same time. They tested their method on 11 real-world datasets and showed it works well in different situations. |
Keywords
» Artificial intelligence » Regression
