Track
ai-learning
AI Learning
36 lessons: how models learn from data—loss, splits, overfitting, leakage, and study habits—plus 108 MCQs.
- Mode
- none
- Practice
- Read / quiz
- Lessons
- 36 units
Before you start
How learning works: data as experience, loss, generalization, leakage, baselines, and the habit of changing one thing at a time.
People train models they cannot explain. This track is the reasoning you need before you touch a training loop.
Study plans, experiment reviews, and the first weeks of any ML project.
Work one idea per lesson against a dataset you understand. Hold out data. Keep a log. Answer the MCQs.
After /data-science/intro and /python/intro—before or alongside /train-models/intro.
Lesson order
Sequential — follow top to bottom36 lessons are live in this track. Start from step 01 for the smoothest path.
-
01 intro What learning means for a model
beginner
Open → -
02 learning-problem The learning problem
beginner
Open → -
03 data-as-experience Data is the experience
beginner
Open → -
04 labels Labels and who makes them
beginner
Open → -
05 loss Loss: the score training reduces
beginner
Open → -
06 generalization Generalization to new cases
beginner
Open → -
07 overfitting Overfitting
beginner
Open → -
08 underfitting Underfitting
beginner
Open → -
09 bias-variance Bias and variance
intermediate
Open → -
10 train-test Train and test
intermediate
Open → -
11 validation Validation for decisions
intermediate
Open → -
12 features Features the model can use
intermediate
Open → -
13 representations Representations
intermediate
Open → -
14 gradients-intuition Gradients without the mythology
intermediate
Open → -
15 learning-rate Learning rate
intermediate
Open → -
16 epochs Epochs and passes over the data
intermediate
Open → -
17 batches Batches
intermediate
Open → -
18 regularization Regularization
intermediate
Open → -
19 early-stopping Early stopping
intermediate
Open → -
20 metrics Metrics that match the decision
intermediate
Open → -
21 baselines Baselines
intermediate
Open → -
22 error-analysis Error analysis
intermediate
Open → -
23 data-quality Data quality before models
intermediate
Open → -
24 leakage Leakage
advanced
Open → -
25 experiment-log Experiment logs
advanced
Open → -
26 reproducibility Reproducibility
advanced
Open → -
27 transfer Transfer from another task
advanced
Open → -
28 fine-tuning-idea Fine-tuning
advanced
Open → -
29 continual Learning after launch
advanced
Open → -
30 evaluation-habits Evaluation habits
advanced
Open → -
31 human-feedback Learning from people
advanced
Open → -
32 curriculum-learning Ordering the examples
advanced
Open → -
33 study-plan How to study this track
advanced
Open → -
34 project-loop The project loop
advanced
Open → -
35 interview Interview review
advanced
Open → -
36 capstone A complete learning report
advanced
Open →