Abstract
We present and analyze a sampling algorithm for the basic linear-algebraic problem of ℓ 2 regression. The ℓ 2 regression (or least-squares fit) problem takes as input a matrix A ∈ ℝ n×d (where we assume n ≫ d) and a target vector b ∈ ℝ n, and it returns as output cross Z sign = min x∈ℝd |b - Ax| 2. Also of interest is x opt = A +b, where A + is the Moore-Penrose generalized inverse, which is the minimum-length vector achieving the minimum. Our algorithm randomly samples r rows from the matrix A and vector b to construct an induced ℓ 2 regression problem with many fewer rows, but with the same number of columns. A crucial feature of the algorithm is the nonuniform sampling probabilities. These probabilities depend in a sophisticated manner on the lengths, i.e., the Euclidean norms, of the rows of the left singular vectors of A and the manner in which b lies in the complement of the column space of A. Under appropriate assumptions, we show relative error approximations for both cross Z sign and x opt. Applications of this sampling methodology are briefly discussed.
Original language | English (US) |
---|---|
Pages | 1127-1136 |
Number of pages | 10 |
DOIs | |
State | Published - 2006 |
Event | Seventeenth Annual ACM-SIAM Symposium on Discrete Algorithms - Miami, FL, United States Duration: Jan 22 2006 → Jan 24 2006 |
Other
Other | Seventeenth Annual ACM-SIAM Symposium on Discrete Algorithms |
---|---|
Country/Territory | United States |
City | Miami, FL |
Period | 1/22/06 → 1/24/06 |
ASJC Scopus subject areas
- Software
- General Mathematics