TheoremComplete

Banach Fixed-Point Theorem

The Banach Fixed-Point Theorem (also called the Contraction Mapping Theorem) guarantees that every contraction on a complete metric space has a unique fixed point. This powerful result is the foundation for proving existence and uniqueness of solutions to differential equations, integral equations, and iterative algorithms.


Statement

Theorem8.1Banach Fixed-Point Theorem

Let (X,d)(X, d) be a complete metric space and f:Xβ†’Xf : X \to X a contraction mapping: there exists 0≀λ<10 \leq \lambda < 1 such that

d(f(x),f(y))≀λd(x,y)forΒ allΒ x,y∈X.d(f(x), f(y)) \leq \lambda d(x, y) \quad \text{for all } x, y \in X.

Then:

  1. ff has a unique fixed point xβˆ—βˆˆXx^* \in X: f(xβˆ—)=xβˆ—f(x^*) = x^*.
  2. For any x0∈Xx_0 \in X, the sequence xn+1=f(xn)x_{n+1} = f(x_n) converges to xβˆ—x^*.
  3. The error satisfies d(xn,xβˆ—)≀λn1βˆ’Ξ»d(f(x0),x0)d(x_n, x^*) \leq \frac{\lambda^n}{1 - \lambda} d(f(x_0), x_0).

Proof

Proof

Existence: Fix x0∈Xx_0 \in X and define xn+1=f(xn)x_{n+1} = f(x_n). We show (xn)(x_n) is Cauchy.

For m>nm > n,

d(xm,xn)β‰€βˆ‘k=nmβˆ’1d(xk+1,xk)=βˆ‘k=nmβˆ’1d(f(xk),f(xkβˆ’1))β‰€βˆ‘k=nmβˆ’1Ξ»kd(x1,x0)≀λn1βˆ’Ξ»d(x1,x0).d(x_m, x_n) \leq \sum_{k=n}^{m-1} d(x_{k+1}, x_k) = \sum_{k=n}^{m-1} d(f(x_k), f(x_{k-1})) \leq \sum_{k=n}^{m-1} \lambda^k d(x_1, x_0) \leq \frac{\lambda^n}{1 - \lambda} d(x_1, x_0).

As nβ†’βˆžn \to \infty, d(xm,xn)β†’0d(x_m, x_n) \to 0, so (xn)(x_n) is Cauchy. By completeness, xnβ†’xβˆ—x_n \to x^* for some xβˆ—βˆˆXx^* \in X. By continuity of ff,

f(xβˆ—)=f(lim⁑xn)=lim⁑f(xn)=lim⁑xn+1=xβˆ—.f(x^*) = f(\lim x_n) = \lim f(x_n) = \lim x_{n+1} = x^*.

Uniqueness: If f(y)=yf(y) = y, then d(xβˆ—,y)=d(f(xβˆ—),f(y))≀λd(xβˆ—,y)d(x^*, y) = d(f(x^*), f(y)) \leq \lambda d(x^*, y). Since Ξ»<1\lambda < 1, this implies d(xβˆ—,y)=0d(x^*, y) = 0, so xβˆ—=yx^* = y.

β– 

Applications

ExamplePicard-LindelΓΆf theorem (ODEs)

The initial value problem yβ€²=f(t,y)y' = f(t, y), y(t0)=y0y(t_0) = y_0 has a unique solution if ff is Lipschitz in yy. Proof: rewrite as y(t)=y0+∫t0tf(s,y(s)) dsy(t) = y_0 + \int_{t_0}^t f(s, y(s)) \, ds and show the integral operator is a contraction on C([t0,t1])C([t_0, t_1]) with appropriate norm.

ExampleComputing √2 via iteration

Define f(x)=(x+2/x)/2f(x) = (x + 2/x)/2 on [1,2][1, 2]. This is a contraction (with small enough Ξ»\lambda), and its fixed point satisfies x=(x+2/x)/2x = (x + 2/x)/2, i.e., x2=2x^2 = 2. Starting with x0=1x_0 = 1, the iteration xn+1=(xn+2/xn)/2x_{n+1} = (x_n + 2/x_n)/2 converges to 2\sqrt{2} (Newton's method).


Summary

The Banach Fixed-Point Theorem is a cornerstone of analysis:

  • Contractions on complete spaces have unique fixed points.
  • Provides constructive proof via iteration.
  • Applications: ODEs, PDEs, integral equations, numerical methods.

See Completeness for background.