9/14/22

ESQC Mathematics
Lecture 2
By Simen Kvaal
[email protected]
Drop me a
message!

Matrices
We pick up from last time

1
 9/14/22

Quick recap: Linear operator as matrix

Space of matrices
• A matrix is a table
# cols
# rows

• Vectors are also matrices!

2
 9/14/22

Matrix—matrix product
• C(x) = A(B(x)) is a linear operator.
Definition : Matrix product

A ∈ M(n, m, F) B ∈ M(m, o, F) C = AB ∈ M(n, o; F)

n
!
Cik = Ai j B jk .
j=1

A(BC) = (AB)C

(A + B)C = AC + BC A(B + C) = AB + AC

AB ! BA

Computing the matrix product

Also, since x is a matrix, we write

3
 9/14/22

Important matrix operations

• Transpose:

• Hermitian adjoint:

• Inner product as matrix product:

General finite-dimensional
vector spaces
With several examples

4
 9/14/22

looks

Space of polynomials like a

basis!
dimension

• Polynomials of degree ≤ n
• A function space we see that
we get a
• Show: Jupyter notebook
matrix!
• Differentiation operator (n = 4)

Space of matrices
• The space M(n) of square matrices (over some field) is a vector space

• It is equal to Euclidean space in n2 dimensions

• But we have an additional structure:

• Vector space with vector-vector multiplication = algebra

5
 9/14/22

A finite-dimensional C*-algebra
• In the second-quant lectures,
A very important
example from the
• We can consider an operator which is a polynomial mathematical
point of view

•
If N spin-orbitals: max N particles, so max degree is 2N
• So a finite dimensional vector space X of operators
• A vector space with a multiplication operation

Inner product, norm

• What these examples lack compared to Euclidean space:
• A sense of distance
• Euclidean space, as model of reality, comes with the intuition of which points
are close to each other

6
 9/14/22

General vector spaces

Vector space
(linear
Mathematical abstraction
Euclidean space structure)
Comes with
linear structure +
inner product

Topology, Other
e.g., inner structures,
product, norm, e.g.,
or metric multiplication

Definition : Vector space

F V + : V×V →
V · : F×V → V x y z∈V α, β ∈ F

😬
0∈V 0+x= x x∈V

x + (y + z) = (x + y) + z

x+y=y+x

x$ x + x$ = 0

(αβ) · x = α · (β · x)

1·x= x

(α + β) · x = α · x + β · x

α · (x + y) = α · x + α · y

7
 9/14/22

Definition : Linear independence, dimension

V L⊂V L
{vi | 1 ≤ i ≤ k} ⊂ L
k
!
ai vi = 0 =⇒ ai = 0 i
i=1

V V

• In Euclidean space: the standard basis

• Polynomials: the various xi

Basis for finite-dimensional spaces

Definition : Basis

V n
{b1 , · · · , bn } n

Theorem

B = {b1 , · · · , bn } V dim(V) < +∞ v∈V

n
!
v= vi bi .
i=1

8
 9/14/22

Example
• The standard basis in Euclidean space:

• The monomials in polynomial space:

• A basis is never unique

Examples: bases in the plane

• Standard basis, non-orthogonal basis, and not-a-basis

9
 9/14/22

Examples of infinite dimensions

Approximations
of functions
• The space of all polynomials, unlimited degree
• The space of all sequences

For infinite
basis
• The space of quadratically integrable functions expansions

Quantum
mechanics!

All finite dimensional vector spaces are

isomorphic – the same
• (… when it comes to the linear structure)

Action of
• And linear operators are …. matrices!
operator in the
given basis

10
 9/14/22

Finite-dimensional Hilbert spaces

Definition : Inner product

V !·, ·" : V × V → F

!x, x" ≥ 0 !x, x" = 0 x=0

!x, αy + βz" = α !x, y" + β !x, z"

!αy + βz, x" = ᾱ !y, x" + β̄ !z, x"

!x, y" = !y, x"

• Finite dim vector space + inner product = Hilbert space

All finite-dimensional Hilbert spaces are

the same
• … when an orthonormal basis is selected “overlap
matrix”
• Let V be a finite dim Hilbert space with given basis

Orthonormal
• Inner prod induces an inner product on
basis
• It is not the Euclidean inner product unless

11
 9/14/22

Remark

Fn M(n, m, F)

More on matrices
Matrices are very central to finite dimensional spaces

12
 9/14/22

Examples of matrices in 2D Euclidean

space
• Show Jupyter notebook

End of lecture 2
• That’s it for today!

13