Randomized algorithms for low-rank matrix factorizations: sharp performance bounds.
Rafi WittenEmmanuel J. CandèsPublished in: CoRR (2013)
Keyphrases
- randomized algorithms
- matrix factorization
- low rank
- lower bound
- collaborative filtering
- approximation algorithms
- low rank matrix
- factorization methods
- missing data
- recommender systems
- matrix completion
- negative matrix factorization
- data matrix
- worst case
- practical problems
- nonnegative matrix factorization
- upper bound
- randomized algorithm
- tensor factorization
- special case
- stochastic gradient descent
- sparsity constraints
- learning algorithm
- generative model
- feature extraction