[ 4. -0.4 -2. ]
[-6.2 -6.4 2.2]
[-15.8 11.6 20.2]
[ 74.2 34.16 -33.8 ]
[ 33.88 -131.44 -101.48]
[-598.28 -94. 395.32]

E21 Computer Engineering Fundamentals
November 5, 2026
[ 4. -0.4 -2. ]
[-6.2 -6.4 2.2]
[-15.8 11.6 20.2]
[ 74.2 34.16 -33.8 ]
[ 33.88 -131.44 -101.48]
[-598.28 -94. 395.32]

[-0.5 0.3333 -0.25 ]
[0.2917 0.7083 0.0833]
[ 0.0313 0.6736 -0.1042]
[ 0.1337 0.724 -0.0868]
[ 0.1124 0.7323 -0.112 ]
[ 0.1229 0.7414 -0.1162]

An iterative numerical algorithm for \(Ax=b\) can be terminated when the sequence \[x^{(0)} \rightarrow x^{(1)} \rightarrow x^{(2)} \rightarrow ... \rightarrow x^{(k)}\] sufficiently converges
Under certain conditions, this is guaranteed to happen.
Typically, we use while loops to implement this:
Pseudo-code:
while norm(residual) > tolerance:
carry out 1 iteration of numerical method
calculate residualJacobi and Gauss-Seidel methods are guaranteed to converge to a solution of \(Ax=b\) $under either of the following conditions:
The following four statements are equivalent, and any one of them implies all the others.
numpy package contains numpy.linalg.solve(A,b) which pretty much always works if a uniques solution to \(Ax=b\) existsThe need for pivoting

The condition number of a matrix is defined as the product of the norm of the matrix and the norm of its inverse:
\[\text{cond } A = || A || \cdot || A^{-1}||\]
Note: When computing \(|| A ||\) for a matrix we will use the Frobenius norm: \[|| A || \equiv \sqrt{\sum_{i=1}^{N} \sum_{j=1}^N A_{ij}^2}.\]
Ill-conditioned matrices lead to error accumulation
| Description | Criterion |
|---|---|
| Well conditioned | \(\text{cond}(A) \approx 1\) |
| Ill conditioned | \(\text{cond}(A) \gg 1\) |
| Singular | \(\text{cond}(A) \rightarrow \infty\) |
Heuristic:
If a matrix is ill-conditioned, its determinant (or the determinant of its inverse) is very small.
But what does ‘small’ mean?
A matrix’s determinant is very small if \[\det A \ll ||A||\]
E21 • Fall 2026 • Lecture 18 • November 5, 2026 • ↩︎