Reinforcement learning
How reinforcement learning agents train through interaction, and why controlling the environment or reward signal is all an adversary needs.
Anomaly detection
How One-Class SVM, Isolation Forest, and LOF work, why their training data is the real attack surface, and what red teamers gain from understanding an anomaly.
K-means clustering
How K-means clustering works, why its centroid mechanics are trivially exploitable, and what red teamers need to know about attacking unsupervised ML models.
Unsupervised learning
How red teamers exploit unsupervised learning models: poisoning, mimicry, and baseline drift attacks against clustering and anomaly detection.
The AI glossary
A practitioner-first AI glossary with opinions. Every term explained by what it does in production, not what it says on the conference slide.
Support vector machines (SVM)
How support vector machines work, why their margin is an attack surface, and what red teamers need to know about poisoning and evading SVM classifiers.
Naive Bayes
How GoodWords attacks exploit the independence assumption in Naive Bayes classifiers, and what red teamers need to know about evading Bayesian security filters.
Decision trees
How decision trees encode their entire logic as readable rules, and why that makes them uniquely exploitable. Part five of the AI red teaming series.
Logistic regression
How logistic regression works, why it is the most common classifier in security systems, and how red teamers exploit its linear decision boundary in practice.