Linear Algebra
10.36 min read

Functions of One Variable

Before tackling multivariate calculus, we build intuition with functions of one variable f:RRf: \mathbb{R} \to \mathbb{R}. These assign a single number f(x)f(x) to each real number xx. Familiar examples: f(x)=x2f(x) = x^2 (a parabola), f(x)=sin(x)f(x) = \sin(x), f(x)=exf(x) = e^x.

A function can be visualized as its graph: the set of points (x,f(x))(x, f(x)). The graph gives a complete picture of ff. Key features: where is ff large or small? Is it increasing (going up) or decreasing (going down)? Does it have peaks, troughs, or flat regions? All these can be read from the graph.

Defining properties like limits and derivatives for univariate functions first lets us build intuition before generalizing to Rn\mathbb{R}^n.

Formal View

Definition 10.2 — Univariate Function
A function of one variable is a map f:DRf: D \to \mathbb{R} where DRD \subseteq \mathbb{R} is the domain. The graph of ff is the set {(x,f(x)):xD}R2\{(x, f(x)) : x \in D\} \subset \mathbb{R}^2.

Why This Matters

Understanding univariate functions deeply is the foundation for multivariate calculus.

  • Loss curves during training: L(t)\mathcal{L}(t) as a function of training step tt.
  • Activation functions in neural networks: ReLU(x)=max(0,x)\text{ReLU}(x) = \max(0,x), σ(x)=1/(1+ex)\sigma(x) = 1/(1+e^{-x}).
  • Interest rate models: P(t)=P0ertP(t) = P_0 e^{rt} as a function of time.

Quiz

Question 1

The graph of f(x)=x2f(x) = x^2 is:

Question 2

Every subset of R2\mathbb{R}^2 is the graph of a function.

Common Mistakes

  • Confusing a function ff with its graph — the function is the rule; the graph is its visual representation.
  • Thinking the domain is always all of R\mathbb{R} — functions like f(x)=xf(x) = \sqrt{x} have restricted domains.