Differentiation and Integration as Theorems

How analysis revisits familiar calculus operations through hypotheses, proof structure, and theorems such as the mean value theorem and the fundamental theorem of calculus.
Modified

April 26, 2026

Keywords

mean value theorem, Rolle’s theorem, fundamental theorem of calculus, Riemann integral, interchange of limits

1 Role

This page is where the real-analysis module revisits calculus from the theorem side.

You already know how derivatives and integrals behave computationally. Here the question changes:

under exactly what hypotheses are those familiar rules justified?

That is the analysis version of calculus.

2 First-Pass Promise

Read this page after Sequences and Series of Functions.

If you stop here, you should still understand:

  • why differentiation and integration are theorem-level operations, not just procedures
  • what Rolle’s theorem and the mean value theorem really say
  • what the fundamental theorem of calculus is certifying
  • why exchanging a limit with a derivative or integral is never automatic

3 Why It Matters

In early calculus, it is easy to think that:

  • derivatives exist because the formula looks smooth
  • integrals exist because the picture looks nice
  • limits can be moved through everything if the functions seem well behaved

Analysis slows that down.

It asks:

  • is the function continuous?
  • is it differentiable on the open interval and continuous on the closed interval?
  • is the approximation uniform enough to justify an interchange?
  • is the object integrable under the actual definition being used?

This matters across the site because modern papers constantly rely on theorem-level calculus:

  • optimization uses the mean value theorem and smoothness inequalities
  • numerical analysis depends on approximation and remainder control
  • probability and PDE-adjacent arguments often exchange limits and integrals
  • learning theory and kernels rely on careful regularity claims

4 Prerequisite Recall

  • continuity and compactness support existence theorems
  • uniform convergence is stronger than pointwise convergence and better for preserving structure
  • single-variable calculus already introduced derivatives, integrals, and Taylor approximation from a computational point of view

5 Intuition

Analysis does not replace calculus. It clarifies why calculus works.

Three big ideas organize this page.

5.1 Rolle and Mean Value Theorems

These theorems convert qualitative information about endpoints into local derivative information inside an interval.

5.2 Fundamental Theorem of Calculus

This theorem says differentiation and integration are deeply linked, but only under the right continuity and integrability assumptions.

5.3 Interchange Warnings

A sequence of functions may converge nicely enough to preserve continuity, but still not nicely enough to justify term-by-term differentiation. Analysis teaches you to ask what mode of convergence the theorem actually needs.

6 Formal Core

Theorem 1 (Rolle’s Theorem) If \(f\) is continuous on \([a,b]\), differentiable on \((a,b)\), and

\[ f(a)=f(b), \]

then there exists \(c \in (a,b)\) such that

\[ f'(c)=0. \]

This theorem says that if a smooth enough function starts and ends at the same height, then somewhere inside it must have a horizontal tangent.

Theorem 2 (Mean Value Theorem) If \(f\) is continuous on \([a,b]\) and differentiable on \((a,b)\), then there exists \(c \in (a,b)\) such that

\[ f'(c)=\frac{f(b)-f(a)}{b-a}. \]

The mean value theorem is one of the most reused theorem templates in all of analysis and optimization.

Theorem 3 (Fundamental Theorem of Calculus) If \(f\) is continuous on \([a,b]\) and

\[ F(x)=\int_a^x f(t)\,dt, \]

then

\[ F'(x)=f(x). \]

Also, if \(F\) is any antiderivative of \(f\), then

\[ \int_a^b f(x)\,dx = F(b)-F(a). \]

This theorem is the bridge between accumulation and local rate of change.

7 Worked Example

Consider the function

\[ f(x)=x^2 \]

on the interval \([1,3]\).

Its average rate of change on the interval is

\[ \frac{f(3)-f(1)}{3-1} =\frac{9-1}{2} =4. \]

The mean value theorem says there exists \(c \in (1,3)\) such that

\[ f'(c)=4. \]

Since

\[ f'(x)=2x, \]

we solve

\[ 2c=4 \quad\Rightarrow\quad c=2. \]

So the theorem is not only abstractly true; in this case we can find the exact point where the instantaneous rate matches the average rate across the interval.

Now define

\[ F(x)=\int_1^x t^2\,dt. \]

By the fundamental theorem of calculus,

\[ F'(x)=x^2. \]

And since an antiderivative of \(t^2\) is

\[ \frac{t^3}{3}, \]

we get

\[ \int_1^3 t^2\,dt = \frac{3^3}{3} - \frac{1^3}{3} = \frac{26}{3}. \]

The key lesson is that both the derivative claim and the integral claim are justified by theorems with explicit assumptions, not by symbol pushing alone.

8 Computation Lens

This page changes how you should read computational formulas.

When you see a derivative or integral inside an argument, ask:

  • what continuity or differentiability assumptions are in force?
  • is the domain closed, open, or compact?
  • is a theorem like MVT or FTC being used implicitly?
  • is someone interchanging limit and derivative or limit and integral without saying why?

That habit pays off in optimization, numerical analysis, and theorem-heavy ML.

9 Application Lens

9.1 Optimization

Smoothness inequalities and descent arguments are often refined versions of mean-value-theorem logic. When a paper talks about gradient-based control of function values, the MVT family is usually nearby.

9.2 Approximation And Numerical Analysis

Taylor remainders, quadrature ideas, and stability arguments all depend on theorem-level control of derivatives and integrals, not just symbolic formulas.

9.3 Probability And Continuous Models

Whenever a proof wants to exchange limit and integral, or differentiate under the integral sign, the real question is whether the necessary hypotheses are actually available.

10 Stop Here For First Pass

If you can now explain:

  • why Rolle’s theorem and the mean value theorem require both continuity and differentiability assumptions
  • why the fundamental theorem of calculus is a theorem about the relationship between rate and accumulation
  • why calculus rules in research proofs are justified statements, not magic syntax
  • why limit-interchange with derivative or integral requires extra care

then this page has done its first-pass job.

11 Go Deeper

The next natural module step is:

For now, the best live neighboring pages are:

12 Sources and Further Reading

Back to top