COURSE 03Intermediate14 IN-DEPTH LESSONS

Machine Learning Core

The core foundations of machine learning: supervised learning, classification, regression, loss functions, decision trees, and generalization.

Start Lesson 1: Classical machine learning models →← All 29 Courses

Course Syllabus & Units

Unit 1

Supervised

2 lessons
1.1L1

Classical machine learning models

**Classical ML** usually means proven model families that work great on tables of features: linear/logistic regression, decision trees, random forests, gradient boosting, k-nearest neighbors, SVMs, and friends.

Read lesson →
1.2L1

Supervised learning

Supervised learning is learning **with an answer key**. You show a model many examples where both the input and the correct output are known. The model adjusts itself to predict outputs on new inputs. Examples: email → spam/not spam; hou...

Read lesson →
Unit 2

Training Loop

1 lessons
2.1L1

Features, loss, and optimization

Three ideas power most learning systems: 1. **Features** — numbers that describe an input (or learned representations). 2. **Loss** — a score for “how wrong” a prediction is. 3. **Optimization** — the process of adjusting the model to ma...

Read lesson →
Unit 3

Generalization

1 lessons
3.1L1

Overfitting and generalization

**Overfitting** means a model looks great on the training examples but fails on new ones—it memorized quirks instead of learning the pattern. **Generalization** means performance holds up on fresh data from the same kind of situation.

Read lesson →
Unit 4

Unsupervised

1 lessons
4.1L1

Unsupervised and self-supervised learning

**Unsupervised learning** finds structure **without** human answer keys: clusters, compression, anomalies. **Self-supervised learning** creates its own practice problems from raw data—like hiding a word and predicting it. Modern language...

Read lesson →
Unit 5

Rl Preview

1 lessons
5.1L1

Reinforcement learning (preview)

**Reinforcement learning (RL)** learns by **trying actions** and receiving **rewards**—not by reading a full answer key for every state. It powers game agents, some robotics, and parts of how chat models are tuned with preference/reward ...

Read lesson →
Unit 6

Metrics

3 lessons
6.1L1

Calibration and reliability

**Calibration** means a model's 80% confidence is right about 80% of the time. **Reliability** diagrams show that relationship. Fluent AI scores are often *not* calibrated.

Read lesson →
6.2L1

Imbalanced classes

When some labels are rare—fraud, defects, disease flags—**class imbalance** makes naive training and accuracy metrics lie.

Read lesson →
6.3L1

Metrics: precision, recall, and ROC

Beyond accuracy: **precision** (of predicted positives, how many were right), **recall** (of real positives, how many you found), and curves like **ROC** that show tradeoffs as you change thresholds.

Read lesson →
Unit 7

Ensembles

1 lessons
7.1L1

Ensembles and boosting (deeper)

**Ensembles** combine multiple models. **Boosting** builds models in sequence, each focusing more on previous mistakes (e.g., gradient boosting families).

Read lesson →
Unit 8

Features

1 lessons
8.1L1

Feature engineering practice

Turning raw fields into **features** models can use: ratios, bins, embeddings, calendars, text vectorizers—guided by domain sense.

Read lesson →
Unit 9

Selection

1 lessons
9.1L1

Model selection and nested CV

**Model selection** chooses among algorithms/hyperparameters using validation. **Nested cross-validation** reduces the optimistic bias from tuning and evaluating on the same splits.

Read lesson →
Unit 10

Trees

1 lessons
10.1L1

Decision trees and forests

**Decision trees** split data with if/then rules. **Random forests** average many trees trained on random slices—strong tabular baselines.

Read lesson →
Unit 11

Linear

1 lessons
11.1L1

Linear and logistic regression

**Linear regression** predicts numbers with a weighted sum. **Logistic regression** predicts class probabilities with a sigmoid on a weighted sum—still a workhorse classifier.

Read lesson →