# experiment-tracking
6 artikelengetagd met “experiment-tracking”
Experiment-tracking-systemen aanvallen
Technieken voor het misbruiken van experiment-tracking-platforms zoals MLflow, Weights & Biases, Neptune en CometML, waaronder data-exfiltratie, metric-manipulatie, experiment-injectie en het benutten van tracking-metadata voor verkenning.
MLflow-beveiligingshardening
Het beveiligen van MLflow-deployments tegen ongeautoriseerde toegang, experimentmanipulatie en vergiftiging van de model-registry.
Beveiliging van experiment-tracking
Security risks in ML experiment tracking systems: what gets logged, what is sensitive, and how tracking platforms become high-value targets for attackers seeking intellectual property and pipeline access.
Lekken van experiment-metadata
How experiment metadata reveals sensitive information: hyperparameters exposing architecture secrets, loss curves revealing training data properties, run names and tags disclosing project intent, and techniques for extracting intelligence from ML experiment logs.
Aanvalsoppervlak van Weights & Biases
Security analysis of Weights & Biases (W&B/wandb): API key exposure, experiment data leakage, team boundary violations, artifact poisoning, and attack techniques specific to the W&B platform.
Beveiliging van ML-experiment-tracking
Securing experiment tracking systems like MLflow, Weights & Biases, and Neptune.