Uniform convergence means one error bound for the whole domain

Key idea

The sequence \(f_n:E\to\mathbb{R}\) converges uniformly to \(f\) if for every \(\varepsilon>0\), there is \(N\) such that \(n\ge N\) implies \(|f_n(x)-f(x)|<\varepsilon\) for every \(x\in E\). Equivalently, \(\|f_n-f\|_\infty\to0\).

Recognition checklist

• First identify the pointwise limit \(f\).
• Compute or bound the error \(|f_n(x)-f(x)|\).
• Take the supremum over the whole domain.
• If the supremum tends to \(0\), convergence is uniform.
• If the supremum stays away from \(0\), or is infinite, convergence is not uniform.

Worked example

Try it

The domain often decides whether convergence is uniform

Key idea

Uniform convergence is global. A sequence can converge uniformly on a smaller interval but fail on a larger interval because the worst error moves toward an endpoint or escapes to infinity.

Standard examples

• \(x/n\to0\) uniformly on \([0,1]\), because the largest error is \(1/n\).
• \(x/n\) does not converge uniformly to \(0\) on \([0,\infty)\), because the error is unbounded for fixed \(n\).
• \(x^n\to0\) uniformly on \([0,a]\) for \(0<a<1\), because the largest error is \(a^n\).
• \(x^n\) does not converge uniformly to its pointwise limit on \([0,1]\), since continuous functions would have a continuous uniform limit.

Worked example

Try it

Uniform convergence can be tested without knowing the limit first

Key idea

A sequence \(f_n\) is uniformly Cauchy on \(E\) if for every \(\varepsilon>0\), there is \(N\) such that \(m,n\ge N\) implies \(\sup_{x\in E}|f_n(x)-f_m(x)|<\varepsilon\). For real-valued functions, this is equivalent to uniform convergence whenever the pointwise limit exists.

Recognition checklist

• Estimate \(|f_n(x)-f_m(x)|\) without choosing a special \(x\).
• Take the supremum over \(E\).
• Force that supremum below \(\varepsilon\) for all large \(m,n\).
• For function series, apply the same idea to tails of partial sums.

Worked example

Try it

Uniform convergence of series comes from uniform tail control

Key idea

For a series \(\sum u_n(x)\), uniform convergence means the partial sums converge uniformly. The Weierstrass M-test says: if \(|u_n(x)|\le M_n\) for every \(x\) and \(\sum M_n\) converges, then \(\sum u_n\) converges uniformly and absolutely.

M-test steps

• Find a bound \(|u_n(x)|\le M_n\) that does not depend on \(x\).
• Check that \(\sum M_n\) converges as a numerical series.
• Conclude uniform absolute convergence of \(\sum u_n(x)\).
• Use tail estimates \(\sum_{n>N}M_n\) when you need an explicit error bound.

Worked example

Try it

Uniform limits preserve structure that pointwise limits may lose

Key idea

Uniform convergence lets one error bound work near every point, so it is strong enough to preserve continuity. It also behaves well under addition, subtraction, multiplication by a bounded function, and preservation of a common Lipschitz constant.

Preserved properties

• If each \(f_n\) is continuous and \(f_n\to f\) uniformly, then \(f\) is continuous.
• If \(f_n\to f\) and \(g_n\to g\) uniformly, then \(f_n-g_n\to f-g\) uniformly.
• If \(g\) is bounded and \(f_n\to f\) uniformly, then \(g f_n\to g f\) uniformly.
• If each \(f_n\ge0\), then the pointwise limit \(f\ge0\).
• Uniform limits of bounded functions are bounded if one sufficiently late \(f_n\) is bounded.
• If every \(f_n\) is Lipschitz with the same constant \(L\), then the limit is Lipschitz with constant \(L\).

Worked example

Try it

Limits commute with integrals more easily than with derivatives

Key idea

If \(f_n\to f\) uniformly on \([a,b]\), then \(\int_a^b f_n\to\int_a^b f\) because the integral of the error is at most \((b-a)\|f_n-f\|_\infty\). Derivatives are stricter: uniform convergence of \(f_n\) alone does not imply \(f_n^\prime\to f^\prime\).

Theorem patterns

• Integral rule: uniform convergence on a bounded interval allows \(\lim_n\int_a^b f_n=\int_a^b f\).
• Derivative rule: if \(f_n^\prime\to g\) uniformly and \(f_n(x_0)\) converges at one point, then \(f_n\) converges uniformly to a differentiable \(f\) with \(f^\prime=g\).
• Warning: differentiating a pointwise or merely uniform limit term-by-term requires more than uniform convergence of the functions.

Worked example

Try it

Avoid the common uniform convergence traps

Common traps

• Pointwise does not imply uniform: \(x^n\) on \([0,1]\) is the standard warning.
• Uniform depends on the domain: \(x/n\) is fine on \([0,1]\) but fails on \([0,\infty)\).
• Continuity preservation works only for uniform limits: pointwise limits can be discontinuous.
• M-test is sufficient, not necessary: failure to find an \(M_n\) does not automatically prove nonuniform convergence.
• Derivative interchange needs derivative control: uniform convergence of \(f_n\) alone is not enough.
• Series terms must vanish uniformly: if \(\sum u_n\) converges uniformly, then \(u_n\to0\) uniformly.

Worked example

Try it

Final recap

• Uniform convergence means \(\sup_x|f_n-f|\to0\).
• Uniform convergence always implies pointwise convergence, but not conversely.
• The same formula can be uniform on one domain and fail on a larger domain.
• The uniform Cauchy criterion controls \(\sup_x|f_n-f_m|\).
• The M-test proves uniform absolute convergence of function series.
• Uniform limits of continuous functions are continuous.
• Uniform convergence also preserves boundedness, nonnegativity, bounded multiplication, and shared Lipschitz constants under the usual hypotheses.
• Uniform convergence on \([a,b]\) lets limits pass through definite integrals.
• Passing limits through derivatives requires uniform convergence of derivatives plus convergence at one point.