ICA vs PCA — the key difference.
hardAnswer
- PCA: finds uncorrelated components maximizing variance.
- ICA: finds statistically independent components — stronger requirement.
- Uses non-Gaussianity (kurtosis, negentropy) to find directions that look non-Gaussian.
- Classic use: blind source separation (cocktail party — separate mixed audio signals).
- Requires ≤ and at most one Gaussian source.
- FastICA is the standard implementation.
Check yourself — multiple choice
- Same thing
- PCA: uncorrelated + max variance; ICA: statistically independent via non-Gaussianity; classic use = cocktail-party source separation; needs ≤ 1 Gaussian source
- Random
- Only linear
ICA: statistical independence via non-Gaussianity; source separation.
#dimensionality-reduction
Practise Unsupervised Learning
214 interview questions in this topic.