Dimensionality Reduction Quiz
Check your understanding of PCA and its neighbors — what the components are, how much variance they explain, why standardization matters, and what t-SNE plots do (and don't) tell you.
Exercise
What is the first principal component?
Need a hint?
Project the data onto a line — which line keeps the data most spread out?
Exercise
Mathematically, the principal components are the…
Need a hint?
PCA diagonalizes a particular matrix built from the data.
Exercise
The 'explained variance ratio' of a principal component equals…
Need a hint?
Each eigenvalue measures variance along its component.
Exercise
Why standardize features before PCA?
Need a hint?
Variance has units. Income in dollars has huge variance; age in years doesn't.
Exercise
What does t-SNE preserve that makes it good for visualization — and what does it sacrifice?
Need a hint?
Trust the clusters; don't measure with a ruler between them.
Exercise
Running PCA to reduce 500 features to 20 before training a classifier changes which slot of the ML project loop?
Need a hint?
It reshapes the inputs the model sees, before any model or loss is chosen.
Exercise
Transfer test — a technique this course did NOT teach: a learning-rate warmup-then-decay schedule. Which slot of the ML project loop does it modify?
Need a hint?
It governs how the parameter search proceeds over training.