Summary of Survattack: Black-box Attack on Survival Models Through Ontology-informed Ehr Perturbation, by Mohsen Nayebi Kerdabadi et al.
SurvAttack: Black-Box Attack On Survival Models through Ontology-Informed EHR Perturbation
by Mohsen Nayebi Kerdabadi, Arya Hadizadeh Moghaddam, Bin Liu, Mei Liu, Zijun Yao
First submitted to arxiv on: 24 Dec 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
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 introduces SurvAttack, a novel black-box adversarial attack framework that leverages clinically compatible and semantically consistent perturbations on longitudinal electronic health records (EHRs) to degrade the predictive performance of state-of-the-art survival analysis models. The proposed algorithm manipulates medical codes throughout a patient’s history using a greedy approach, prioritized by a composite scoring strategy considering saliency, stealthiness, and clinical meaningfulness. The authors demonstrate the effectiveness of SurvAttack in illustrating the vulnerabilities of patient survival models, model interpretation, and ultimately contributing to healthcare quality. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about making sure that medical predictions are safe from being tricked or fooled by fake information. It introduces a new way to test how well these predictions work by adding tiny changes to a person’s medical history. This helps doctors understand what would happen if they made certain decisions, and it could improve healthcare quality. |
