Algebraic Structures and Formal Power Series Expansion

Algebraic Context: The study of polynomial expansion and manipulation originates from algebra. A set such as $R[x]$ functions as both a ring and a vector space.
Expansion Bases: When expanding a function, one is not limited to expanding in terms of $x$ . For example, an expansion can be performed in terms of $(x - 4)$ or $(x + 1)$ . The choice of basis depends on the specific algebraic problem.
Linear Algebra Connection: The space of polynomials is treated as a vector space over a ring $R$ . This allows for the use of basis elements and coordinate representations.

Degree Properties of Polyonmials:
- The degree of a constant is $0$ .
- If a polynomial $g(x)$ has a degree greater than zero, multiplying it by another polynomial results in the addition of their degrees: $deg(f(x) \times g(x)) = deg(f(x)) + deg(g(x))$ .
- The degree of a polynomial is either $-\times$ (for the zero polynomial) or an integer $\ge 0$ .
- If a product of polynomials results in a degree of $0$ , the components cannot have a degree of $-\infty$ . The degree must be at least $0$ .
Inverses in $R[x]$ :
- Constants within the ring $R$ often have inverses. For example, if $2$ is in $R$ , then the constant polynomial $\frac{1}{2}$ is in $R[x]$ . Similarly, $\frac{1}{3}$ may be in $R[x]$ if $3$ has an inverse in $R$ .
- The zero element ( $0$ ) does not have an inverse.
- Crucially, while division can be performed within the coefficient ring $R$ , division with respect to the variable $x$ is generally not possible in the polynomial ring $R[x]$ .
Composition of Functions: Given two polynomials $f(x)$ and $g(x)$ , one can form the composition $f(g(x))$ .
- Example: If $f(x) = x^2 + x - 1$ , the composition would substitute $g(x)$ into every instance of $x$ within $f$ .

Definition: A formal power series is treated as an infinite sequence of coefficients. While it is written in the form of a power series, it is "formal" because convergence is not an immediate concern; the focus is on the manipulation of the sequences themselves.
Multiplication (Cauchy Product): To multiply two formal power series, the coefficients are determined term by term:
- Constant Term: Obtained solely by the product of the two constant terms: $a_0 \times b_0$ .
- Linear (First Order) Term ( $x$ ): Obtained from $(a_0 \times b_1) + (a_1 \times b_0)$ . For example, if the coefficients result in $2x$ , that is the linear component.
- Quadratic Term ( $x^2$ ): Obtained from $(a_0 \times b_2) + (a_1 \times b_1) + (a_2 \times b_0)$ . In a specific example provided, this results in $3x^2$ .
- Cubic Term ( $x^3$ ): Obtained from $(a_0 \times b_3) + (a_1 \times b_2) + (a_2 \times b_1) + (a_3 \times b_0)$ .
Invertibility in Formal Power Series:
- Items that do not appear to be invertible in standard polynomial rings may be invertible in formal power series.
- Example: The expression $(1 - x)$ is invertible in the context of formal power series, whereas $\frac{1}{1-x}$ is not a polynomial.
- Determining the inverse is done term by term (low coefficient computation) without needing to worry about higher-order terms during the calculation.

Formal vs. Analytical: "Formal" means only the coefficients matter. However, if the series converges, it becomes a functional power series.
Pointwise Convergence: When writing $1 + x + x^2 + x^3 + \dots = \frac{1}{1 - x}$ , we mean that for any $x$ where the absolute value is smaller than one ( $|x| < 1$ ), the series converges to the value of the function. This is a statement of pointwise convergence.
Types of Expansions and Transforms:
- Power Series: Expanding a function $f(x)$ into a sum of monomials weighted by coefficients.
- Fourier Series: Expanding a function with a specific period (e.g., $0$ to $2\pi$ ) into a combination of trigonometric functions.
- Laurent Series: A more general form of power series including negative powers.
- Lagarderex Series: Another specific expansion type mentioned in relation to finding coefficients.
Calculus operations: If a series converges within a reasonable range, it can be differentiated term by term. Substituting values (e.g., substituting $-x$ for $x$ ) is also valid within the convergence range.
Logarithmic Expansion: By substitution and integration, one can derive expansions such as:
- $\ln(1 - x) = -x - \frac{x^2}{2} - \dots$

Relationship between Inverses and Squaring:
- Question: Is the inverse squared always the identity, or when is that true?
- Response: The speaker clarifies the algebraic manipulation of inverses. If $f(x) = g(x)^2$ , then the inverse of $f(x)$ is the square of the inverse of $g(x)$ . Written as: $f(x)^{-1} = (g(x)^2)^{-1} = (g(x)^{-1})^2$ .
- Example: If $g(x)^{-1} = h(x) = (1 - x)$ , then $f(x)^{-1} = h(x)^2 = (1 - x)^2$ , which expands to $1 - 2x + x^2$ .
The Geometric Series:
- Question: For $1 + x$ , looking at $s = 1 + x + x^2 + x^3 + \dots$ , if you multiply both sides by $(1 - x)$ , the right side equals $1$ . Isn't that just $\frac{1}{1 - x}$ ?
- Response: Yes. In terms of formal power series, $1 + x + x^2 + \dots$ is explicitly the inverse of $(1 - x)$ . We use the notation $(1 - x)^{-1}$ . This holds as a functional equality as long as the absolute value of $x$ is less than $1$ .