# The Hidden Adversarial Vulnerabilities of Medical Federated Learning

arXiv: https://arxiv.org/abs/2310.13893 · HTML: https://mlsanity.com/paper/2310.13893/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 16 references that are in the map.
- First posted on arXiv 21 Oct 2023.

## Builds on (most cited first, 11 of 11)

- [Practical Black-Box Attacks against Machine Learning](https://mlsanity.com/paper/1602.02697/) (2016)
- [On Evaluating Adversarial Robustness](https://mlsanity.com/paper/1902.06705/) (2019)
- [Rotation Equivariant CNNs for Digital Pathology](https://mlsanity.com/paper/1806.03962/) (2018)
- [You Only Propagate Once: Accelerating Adversarial Training via Maximal Principle](https://mlsanity.com/paper/1905.00877/) (2019)
- [Understanding Adversarial Attacks on Deep Learning Based Medical Image Analysis Systems](https://mlsanity.com/paper/1907.10456/) (2019)
- [8-Bit Approximations for Parallelism in Deep Learning](https://mlsanity.com/paper/1511.04561/) (2015)
- [Efficient Adversarial Training with Transferable Adversarial Examples](https://mlsanity.com/paper/1912.11969/) (2019)
- [Adversarial Attack Vulnerability of Medical Image Analysis Systems: Unexplored Factors](https://mlsanity.com/paper/2006.06356/) (2020)
- [Exploring adversarial attacks in federated learning for medical imaging](https://mlsanity.com/paper/2310.06227/) (2023)
- [A Comparative Study of Federated Learning Models for COVID-19 Detection](https://mlsanity.com/paper/2303.16141/) (2023)
- [AI Technical Considerations: Data Storage, Cloud usage and AI Pipeline](https://mlsanity.com/paper/2201.08356/) (2022)

## Built on by (most cited first, 1 of 1)

- [Threats and Defenses in Federated Learning Life Cycle: A Comprehensive Survey and Challenges](https://mlsanity.com/paper/2407.06754/) (2024)

## Abstract (arXiv)

In this paper, we delve into the susceptibility of federated medical image analysis systems to adversarial attacks. Our analysis uncovers a novel exploitation avenue: using gradient information from prior global model updates, adversaries can enhance the efficiency and transferability of their attacks. Specifically, we demonstrate that single-step attacks (e.g. FGSM), when aptly initialized, can outperform the efficiency of their iterative counterparts but with reduced computational demand. Our findings underscore the need to revisit our understanding of AI security in federated healthcare settings.