Machine Learning Associate — Illustrated Study Guide

A practical, illustrated study guide for the Machine Learning Associate certification. Five modules take you through the full classical-ML toolkit: the ML workflow and the supervised/unsupervised divide; the major supervised algorithms; unsupervised learning and dimensionality reduction; the bias–variance tradeoff with overfitting and regularization; and honest evaluation with the right metrics. Rich with Mermaid diagrams, algorithm cheat-sheets, worked numeric examples, key takeaways, and self-checks.

Key facts

• Course price: $91 (one-time payment)
• Field: Artificial Intelligence & LLMs
• Prepares for: Machine Learning Associate
• 5 modules · 15 lessons · about 4 hours of lessons
• Self-paced, online, illustrated lessons

Course outline

1. The ML Workflow & Learning Paradigms — The end-to-end machine-learning workflow and the supervised vs unsupervised divide that frames every project.
   • The End-to-End Machine Learning Workflow
   • Supervised vs Unsupervised Learning
   • Train, Validation, and Test Splits
2. Supervised Algorithms — Linear and logistic regression, decision trees, ensembles (random forests, gradient boosting), kNN, SVM, and naive Bayes.
   • Linear and Logistic Regression
   • Trees, Random Forests, and Gradient Boosting
   • kNN, SVM, and Naive Bayes
3. Unsupervised Learning & Dimensionality Reduction — Clustering with k-means and hierarchical methods, plus dimensionality reduction with PCA.
   • k-Means Clustering
   • Hierarchical Clustering
   • PCA and Dimensionality Reduction
4. Bias, Variance, and Regularization — The bias–variance tradeoff, diagnosing overfitting and underfitting, and taming complexity with L1/L2 regularization and cross-validation.
   • The Bias-Variance Tradeoff
   • Overfitting, Underfitting, and Regularization
   • Cross-Validation Done Right
5. Evaluation Metrics & Real-World Pitfalls — Choosing the right metric — accuracy, precision, recall, F1, ROC-AUC, RMSE/MAE — and avoiding class imbalance and data leakage traps.
   • Classification Metrics and the Confusion Matrix
   • ROC-AUC and Regression Metrics
   • Class Imbalance and Data Leakage