Sparse coding — what is it?
hardAnswer
- Learn overcomplete dictionary D such that x ≈ Dα with sparse α (few non-zero entries).
- Uses: image compression, denoising, feature learning.
- Fit alternates between (1) LASSO for α given D, (2) dictionary update for D given α.
- Foundation of Olshausen & Field's V1 modeling and later dictionary learning.
- Related to compressed sensing.
Check yourself — multiple choice
- Random
- Overcomplete dictionary D + sparse α such that x ≈ Dα; alternates LASSO on α + D update; V1 modeling / compressed sensing
- Same as PCA
- Not real
Sparse coding: overcomplete dictionary + sparse α; V1 / compressed sensing.
#dimensionality-reduction#representation-learning
Practise Unsupervised Learning
214 interview questions in this topic.