Matrix completion — how does it relate to unsupervised learning?
hardAnswer
- Recover missing entries in a matrix by assuming low-rank structure.
- Solve min s.t. observed entries match — NP-hard, relaxed to min ||M||_* (nuclear norm).
- Foundation of collaborative filtering (Netflix Prize), missing-data imputation.
- Modern: Alternating Least Squares (Spark ALS), deep matrix factorization, or neural collaborative filtering.
Check yourself — multiple choice
- Random
- Fill missing entries assuming low-rank M; solved via nuclear-norm relaxation or ALS; foundation of collaborative filtering / imputation
- Same as PCA
- Not real
Matrix completion: low-rank fill-in; collaborative filtering / imputation.
#dimensionality-reduction#applications
Practise Unsupervised Learning
214 interview questions in this topic.