PhysioAttack-FL: A Benchmark for Adversarial Behavior Detection in Federated Learning

A benchmark dataset for detecting adversarial behavior in federated learning systems for the Internet of Things. The dataset was sourced from Kaggle, but details about its creator, size, and specific temporal coverage are not provided in the available metadata.

Use Cases

• Benchmarking adversarial detection algorithms based on the described focus on federated learning IoT systems.
• Training machine learning models to identify malicious clients based on the adversarial behavior patterns the dataset is designed to capture.
• Studying the robustness of federated learning aggregation rules against the types of attacks simulated in the data.

Strengths

• The dataset is explicitly designed as a benchmark, suggesting a structured evaluation purpose.
• It addresses a specific and relevant niche: security for federated learning in IoT.

Limitations

• Row count, file size, and column definitions are unknown, which limits suitability assessment.
• Description metadata is limited; actual data quality and structure require manual inspection after download.
• Last update date and license are unknown; freshness and usage rights are unverified.

Provenance

Source

Kaggle