Deep Learning · Level 2 of 5
Feature Hierarchy
Representations arranged from simpler patterns to more abstract combinations.
Higher layers often combine lower-layer signals, but interpretations are not guaranteed.
Example
Early vision filters detect edges that later stages combine into shapes.
Listen to the definition and example
Audio transcript
Feature Hierarchy. Representations arranged from simpler patterns to more abstract combinations. Higher layers often combine lower-layer signals, but interpretations are not guaranteed. For example: Early vision filters detect edges that later stages combine into shapes.
Explore this concept
Why it matters
This helps explain why a neural model learns well, becomes unstable, or fails to generalize.
Quick recall question
Try answering before looking back at the definition.