How best to re-learn Matrices for a pre-requisite of Linear Algebra

April 6, 2026

Questions and Doubts (Answers from Claude)

Is (CE)ᵀ = EᵀCᵀ always true? Yes, always. Called the reverse order law. Extends to (CDE)ᵀ = EᵀDᵀCᵀ.
Does AᵀA also result in a symmetric matrix like AAᵀ? Yes. Both AAᵀ and AᵀA are always symmetric, just not the same matrix.
Do I need to do determinants for larger than 3×3? No. 2×2 and 3×3 is enough. Larger ones are computed programmatically.
What is the purpose of Gaussian Elimination? To solve systems of linear equations (Ax = b) by simplifying the augmented matrix [A|b] into REF or RREF, from which you read off the unknowns.
What’s the point of all three — Gaussian Elimination, REF, RREF? Gaussian Elimination is the process. REF is the intermediate goal enabling back-substitution. RREF is the final goal where the solution is directly readable. It’s one process with two stopping points.
What is the leading coefficient/leading entry in a matrix? The first non-zero element in a row. Also called the pivot. Same thing, interchangeable terms.
What is a triangular matrix and what is the leading coefficient in it? Square matrix where all elements either above (lower triangular) or below (upper triangular) the diagonal are zero. The leading coefficient in each row is always the diagonal element.
If we replace non-diagonal elements with 0, is it still triangular? Yes. Definition only requires the correct side to be zero. Other elements including non-diagonal ones can be anything including zero.
Is Gauss-Jordan Elimination the same as Gaussian Elimination? Not exactly. Gaussian Elimination stops at REF, then you back-substitute. Gauss-Jordan goes all the way to RREF, no back-substitution needed. Gauss-Jordan is Gaussian taken all the way.
Is upper-triangular the necessary first step in Gauss-Jordan? Yes. You eliminate downward first (upper triangular = REF), then upward to reach RREF. Lower triangular doesn’t fit the process.
What is the significance of pivot position? Determines the rank (number of pivots = rank), whether a unique solution exists, and which variables are basic vs free.
What does “augment” mean in general and in this context? Generally: to extend or append something. In this context: appending the constants column b to matrix A giving [A|b], so row operations apply to both sides simultaneously.
What is the determinant’s role in elimination? If a pivot becomes 0 during elimination, you either swap rows or the matrix is singular (det = 0), meaning no unique solution exists.
What is REF? Row Echelon Form — leading entry of each row is right of the one above, all zero rows at bottom, zeros below each pivot.
What is RREF? Reduced Row Echelon Form — everything in REF plus each pivot = 1 and zeros above and below every pivot.
What is the use of REF and RREF? REF: enables back-substitution, useful for rank and determinant. RREF: solution directly readable, used for finding inverse (augment with I), determining rank and free variables.
What is rank? Number of non-zero rows in REF/RREF, equivalently the number of pivot positions.
Are pivot positions the same in REF and RREF? Yes. Row operations don’t change pivot positions, only what’s around them.

I can’t remove the made with AI footer badge for just this page too much work. And I prefer not to ship any JavaScript.

Journal / How best to re-learn Matrices for a pre-requisite of Linear Algebra

A bunch of YouTube Links

Questions and Doubts (Answers from Claude)