# Machine Learning Fundamentals

- **Part of:** [AI & Machine Learning](/en/tracks/ai-machine-learning) — learning path 2 of 3
- **Length:** 4 chapters · 22 lessons · 12 hours a week
- **Level:** Intermediate

::route
1. [Python & Mathematics for ML](/en/paths/python-mathematics-for-ml) — 4 chapters · 10 hours a week · 1 checkpoint
2. **Machine Learning Fundamentals** — 4 chapters · 12 hours a week · 1 checkpoint
3. [ML Engineering & Deployment](/en/paths/ml-engineering-deployment) — 3 chapters · 8 hours a week · 1 checkpoint
::

::cta https://app.artimend.com/tracks/ai-machine-learning Start this track

## Syllabus

### 1. Supervised Learning: Regression

- 6 lessons
- **Skills:** Linear Regression · Regularization · Feature Engineering · Cross-Validation

### 2. Supervised Learning: Classification

- 6 lessons
- **Skills:** Logistic Regression · Decision Trees · Random Forests · XGBoost · Hyperparameter Tuning

### 3. Unsupervised Learning

- 5 lessons
- **Skills:** k-means Clustering · Hierarchical Clustering · DBSCAN · PCA · Dimensionality Reduction

### 4. Introduction to Deep Learning

- 5 lessons
- **Skills:** Neural Networks · Backpropagation · PyTorch · TensorFlow · Keras
- **Checkpoint:** ML Practitioner Gate

## Checkpoints

To pass each checkpoint you hand in a piece of real work. What each one asks for:

- **ML Practitioner Gate:** Train a classifier on a public dataset, beat a baseline by ≥5% on a held-out test set, document the choices. Train a small neural network from scratch (PyTorch or Keras), reproduce a benchmark within 10% of the published number.