low-rank approximation
Low-rank approximation is a technique used in matrix decomposition, where complex matrices are approximated by lower-rank matrices to simplify computations, reduce dimensionality, or compress data while preserving essential structures.
- Demystifying Language Model Forgetting with Low-rank Example Associations
- Efficient Parametric SVD of Koopman Operator for Stochastic Dynamical Systems
- Fast exact recovery of noisy matrix from few entries: the infinity norm approach
- Guarantees for Alternating Least Squares in Overparameterized Tensor Decompositions
- Spectral Perturbation Bounds for Low-Rank Approximation with Applications to Privacy