Jose's Linear Algebra Times 1
Across
- 4. Difference between observed and estimated values in numerical problems.
- 5. Largest absolute value of the eigenvalues of a matrix.
- 7. Resistance of numerical algorithms to errors during computations.
- 10. A measure of the size or magnitude of a matrix (e.g., Frobenius norm, spectral norm).
- 14. Process of converting vectors into orthogonal vectors, e.g., Gram-Schmidt.
- 16. Method for solving linear systems by transforming the matrix into an upper triangular form.
- 17. An iterative algorithm to solve symmetric positive definite systems.
- 20. A matrix classified as positive or negative definite/semidefinite.
- 21. Matrix with all zeros above or below its diagonal (upper/lower triangular).
- 22. A system where most elements in the coefficient matrix are zero, allowing specialized solving methods.
- 23. A matrix divided into smaller submatrices for operations like block multiplication.
- 24. Symmetric matrix with all positive eigenvalues.
Down
- 1. A measure of the magnitude of a vector (e.g., L1 norm, L2 norm, infinity norm).
- 2. A matrix with a high condition number, indicating numerical instability.
- 3. Measure of how much the solution to a problem changes with changes in input.
- 6. Error caused by approximating real numbers in finite precision arithmetic.
- 8. A matrix where AA^T = A^T A
- 9. Study of changes in solutions of equations or eigenvalues when the problem is slightly altered.
- 11. Sensitivity of solutions to small changes in the coefficients of a system of equations.
- 12. Decomposition of A into lower triangular L and upper triangular U matrices.
- 13. Generalized inverse of a matrix, often computed using SVD.
- 15. Rearranging rows or columns during Gaussian elimination to enhance numerical stability.
- 18. Reducing a matrix's size by eliminating a known eigenvalue and eigenvector.
- 19. Factorization of a positive definite matrix into LL^T, where L is lower triangular.