#concept
Representation/feature learning is >> a set of techniques that allow a system to auto-discover representations needed for feature detection / classification from raw data.
As an analogy for representation learning, we may think of a system for identifying faces. As opposed to manually specifying face-identifying features like “look for 2 eyes, a nose, a mouth at (X,Y) positions”, a representation learning system will >> learn on its own the patterns most useful for recognizing faces.
Give an example of an feature learning algorithm that uses vectors! ? Some
- node2vec
- GloVe