Week 2 - Rank of a matrix, Determinants Flashcards
Rank of a matrix
- The no. of NON-ZERO ROWS in the row-echelon form (REF)
- The no. of leading ones
- bounded by min{m, n} for matrix of size m x n
How to write the solution space for Ax=0 if…
1. r=n
2. r<n
If r=n, Ax=0 has a unique solution
{x∈R^s | Ax=0} = {0}
If r<n, Ax=0 has infinitely many solutions
eg. {x∈R^s | Ax=0} = [sv1 + tv2 | s,t∈R}
For A=m x n matrix, how many linearly independent (LI) vectors are there?
(n - r) LI vectors
where r is the rank
Basis
A set of vectors B where every vector in the {solution} space is a UNIQUE LINEAR COMBINATION of the vectors in B
eg. {v1, v2}
General solution to Ax=0
General solution to Ax=b
Note: Ap=b is a particular solution to Ax=b
^PRINCIPLE OF LINEARITY
To Ax=0, X=rv1 + sv2
To Ax=b, X =p + rv1 + sv2
- If r=n, Ax=b has a unique solution
- If r<n, Ax=b has infinitely many solutions
How to calculate determinant using cofactor expansion along row i / down column j?
- “a” - element in the matrix
- “sign”
- “minor” - determinant of the square (n-1)(n-1) matrix
Effect of row operations on a matrix on the value of its determinant, for using UPPER-TRIANGULAR FORM (ie. all zeroes below main diagonal)
- R2 -> kR2
- R2 interchange R2
- R2 -> R2 + R1
- R2 -> R2 + kR1
- R2 -> kR2 + R1
- R2 -> kR2
|B| = k|A| - R2 interchange R2
|B| = -|A| - R2 -> R2 + R1
|B| = |A| determinant is unchanged! - R2 -> R2 + kR1
|B| = |A| determinant is unchanged! - R2 -> kR2 + R1
|B| = k|A|
“Sneaky trick” to find p
See ex.2
Determinant of a matrix facts
- If A has a row/column of zeroes, then |A|=0
- If A has 2 identical rows/columns, then |A|=0
From ex.2 class,
|B| = 0 means…
1. LD
2. Infinitely many solutions = non-trivial
3. B is NOT invertible
|B| =/= 0 means…
1. LI
2. Only trivial solutions
3. B is invertible