Fixed Point Principles Practice Quiz with a Step-by-Step Interactive Lesson
Use the quiz at the top of the page to practice fixed point principles: solving \(f(x)=x\), recognizing contractions \(d(Tx,Ty)\le k d(x,y)\) with \(k<1\), separating contractions from nonexpansive maps with constant \(1\), applying the Banach theorem on complete metric spaces, understanding Picard iteration \(x_{n+1}=T(x_n)\) and one-step error shrink, checking self-map and completeness hypotheses, using interval and Brouwer-style existence results, and avoiding uniqueness traps. If you need a refresher, open the lesson for mentally followable examples and quick checks.
How this fixed point principles practice works
1. Take the quiz: answer questions about fixed point equations, contractions, Banach theorem, Brouwer existence, and counterexamples.
2. Open the lesson: review the definitions, theorem hypotheses, and short examples before retrying.
3. Retry: return to the quiz and translate each problem into a fixed point equation or theorem checklist.
What you will learn in the fixed point principles lesson
Fixed point equations
Definition: \(x\) is fixed by \(T\) when \(T(x)=x\).
Computation: solve \(f(x)=x\), not \(f(x)=0\), unless the problem has been rewritten.
Examples: affine maps \(x\mapsto ax+b\), constant maps, the identity map with many fixed points, and \(x\mapsto \cos x\).
Contractions and Banach
Contraction: a uniform distance shrink \(d(Tx,Ty)\le k d(x,y)\) with \(k<1\).
Banach theorem: complete metric space plus contraction self-map gives exactly one fixed point.
Iteration and errors: \(x_{n+1}=T(x_n)\) converges to the fixed point from any starting point, and each error is multiplied by at most \(k\).
Hypotheses and failures
Completeness: keeps the Cauchy limit of the iteration inside the space.
Self-map: every iterate must remain in the same space where the theorem is applied.
Sharp trap: Lipschitz constant \(1\) is nonexpansive, but it is not enough for Banach contraction.
Brouwer-style existence
Interval case: every continuous map \([a,b]\to[a,b]\) has a fixed point.
Finite-dimensional case: continuous self-maps of compact convex subsets of \(\mathbb{R}^n\) have fixed points.
Important distinction: Brouwer gives existence, not uniqueness or iteration convergence.
Ready to test the hypotheses?
Return to the quiz and check whether each question needs a fixed point equation, a contraction estimate, a nonexpansive counterexample, Banach theorem, or Brouwer existence.
Loading...
Higher Analysis
Fixed Point Principles Lesson
1 / 8
Lesson overview
Purpose: Build a reliable fixed point toolkit: translate a question into \(T(x)=x\), recognize contraction estimates, know exactly what Banach theorem guarantees, use iteration \(x_{n+1}=T(x_n)\), check completeness and self-map hypotheses, distinguish Brouwer existence from Banach uniqueness, and avoid common fixed point traps.
Success criteria
Define a fixed point as a point \(x\) with \(T(x)=x\).
Solve simple fixed point equations for affine, constant, identity, and elementary maps.
Recognize contractions through \(d(Tx,Ty)\le k d(x,y)\) with a single \(k<1\).
Distinguish contractions from nonexpansive maps whose best Lipschitz constant is \(1\).
State the Banach theorem: a contraction self-map on a complete metric space has exactly one fixed point.
Use the uniqueness proof \(d(p,q)\le k d(p,q)\) for two fixed points \(p,q\).
Explain why Picard iteration converges to the fixed point under Banach hypotheses and why the one-step error shrinks by at most \(k\).
Spot failures caused by incomplete spaces, missing self-map conditions, or Lipschitz constant \(1\).
Use the interval fixed point result and Brouwer theorem for existence without uniqueness.
Choose between Banach and Brouwer by reading the hypotheses first.
Key vocabulary
Fixed point: \(x\in X\) such that \(T(x)=x\).
Self-map: a map \(T:X\to X\).
Contraction: a map with \(d(Tx,Ty)\le k d(x,y)\) for all \(x,y\), where \(0\le k<1\).
Nonexpansive map: a map with Lipschitz constant at most \(1\); every contraction is nonexpansive, but not conversely.
Complete metric space: every Cauchy sequence converges to a point of the space.
Picard iteration: the sequence \(x_{n+1}=T(x_n)\).
Brouwer existence: continuous self-maps of compact convex subsets of \(\mathbb{R}^n\) have at least one fixed point.
Quick pre-check
Pre-check: Which equation asks for a fixed point of \(T\)?
Hint: Fixed means the input is sent back to itself.
A fixed point is found by solving \(T(x)=x\)
Learning goal: Turn a fixed point question into an equation and solve it without confusing fixed points with zeros.
Key idea
For a map \(T:X\to X\), a fixed point is any \(x\in X\) satisfying \(T(x)=x\). On the real line, this is the intersection of the graph \(y=T(x)\) with the diagonal \(y=x\). For affine maps \(T(x)=ax+b\), the equation is \(x=ax+b\), so \((1-a)x=b\) when \(a≠1\).
Examples
\(T(x)=x/2\) has fixed point \(0\).
\(T(x)=(x+1)/2\) has fixed point \(1\).
\(T(x)=2-x\) has fixed point \(1\), but it is not a contraction in the usual metric.
A constant map \(T(x)=c\) has fixed point \(c\) when \(c\in X\), and its distance shrink is \(0\).
The identity map on \([0,1]\) fixes every point, so existence alone does not mean uniqueness.
Fixed point equation
The fixed point equation is not usually the same as \(T(x)=0\). Sometimes one rewrites it as \(F(x)=T(x)-x=0\), but the original condition remains \(T(x)=x\).
Worked example
Example: Find the fixed point of \(T(x)=1+x/4\) on \(\mathbb{R}\).
Solve \(x=1+x/4\). Then \(3x/4=1\), so \(x=4/3\). Checking gives \(T(4/3)=1+1/3=4/3\).
Try it
Try it: What is the fixed point of \(T(x)=2+x/2\) on \(\mathbb{R}\)?
Hint: Set \(x=2+x/2\) and isolate \(x\).
A contraction shrinks every distance by the same factor below \(1\)
Learning goal: Recognize when a map is a contraction and when it merely preserves distances.
Key idea
A map \(T:X\to X\) on a metric space is a contraction if there is a number \(k<1\) such that \(d(Tx,Ty)\le k d(x,y)\) for every pair \(x,y\in X\). The same \(k\) must work globally. A Lipschitz constant equal to \(1\) is only nonexpansive, so Banach theorem does not necessarily apply.
Recognition checklist
Distance shrink: \(d(Tx,Ty)\le k d(x,y)\) with \(k<1\).
Derivative shortcut: on an interval, \(\sup |T'(x)|\le k<1\) proves a contraction in the usual metric.
Constant maps: \(d(Tx,Ty)=0\), so a constant self-map is a contraction with \(k=0\) whenever \(X\) has at least two points.
Translations: \(T(x)=x+1\) preserves distances, so its Lipschitz constant is \(1\).
Reflection: \(T(x)=2-x\) has a fixed point but also has Lipschitz constant \(1\).
Self-map first: before applying a theorem, check \(T(X)\subseteq X\).
Derivative shortcut
If \(T\) is differentiable on an interval and \(|T'(x)|\le k<1\) everywhere, the mean value theorem gives \(|T(x)-T(y)|\le k|x-y|\). Local small derivative at one point is not enough; the bound must hold on the whole interval used.
Worked example
Example: Why is \(T(x)=1+x/4\) a contraction on \(\mathbb{R}\)?
For any \(x,y\), \(|T(x)-T(y)|=|(1+x/4)-(1+y/4)|=|x-y|/4\). Thus the contraction constant can be \(k=1/4\), which is strictly less than \(1\).
Try it
Try it: Which statement proves that \(T\) is a contraction on \(X\)?
Hint: The estimate must compare distances between all image pairs and all original pairs with a strict shrink factor.
Banach theorem gives existence, uniqueness, and convergence
Learning goal: Know the exact hypotheses and conclusions of the contraction mapping principle.
Key idea
If \((X,d)\) is complete and \(T:X\to X\) is a contraction, then \(T\) has exactly one fixed point \(x^*\). Moreover, from any starting point \(x_0\in X\), the iteration \(x_{n+1}=T(x_n)\) converges to \(x^*\), with \(d(Tx,x^*)\le k\,d(x,x^*)\).
What the theorem gives
Existence: iteration produces a Cauchy sequence, and completeness supplies its limit inside \(X\).
Uniqueness: if \(p,q\) are fixed, then \(d(p,q)=d(Tp,Tq)\le k d(p,q)\), forcing \(p=q\).
Convergence: \(x_n\to x^*\) for every starting point \(x_0\in X\).
Error shrink: \(d(Tx,x^*)\le k\,d(x,x^*)\), so each step moves at most a factor \(k\) away from the fixed point.
Continuity: a contraction is Lipschitz, hence continuous; if \(x_{n+1}=T(x_n)\to L\), continuity gives \(T(L)=L\).
Worked example
Example: Apply Banach theorem to \(T(x)=(x+1)/2\) on \(\mathbb{R}\).
The usual metric on \(\mathbb{R}\) is complete, \(T:\mathbb{R}\to\mathbb{R}\), and \(|T(x)-T(y)|=|x-y|/2\). Banach theorem applies. Solving \(x=(x+1)/2\) gives the unique fixed point \(x=1\), and iteration from any \(x_0\) converges to \(1\).
Try it
Try it: For a contraction self-map on a complete metric space, the iteration \(x_{n+1}=T(x_n)\) converges to:
Hint: The theorem gives both a unique target and convergence of the repeated iteration.
The limit must stay in the space where the theorem is used
Learning goal: See why completeness and the self-map condition are not decorative assumptions.
Key idea
The Banach proof builds a Cauchy sequence from iteration. Completeness is what turns that Cauchy sequence into a point of \(X\). The self-map condition \(T:X\to X\) keeps all iterates in the same space. If either condition fails, the fixed point can lie outside the space or iteration can leave the domain.
Hypotheses that matter
An incomplete space can contain Cauchy sequences whose limits are missing.
A contraction on an incomplete space may have no fixed point in that space.
A map that is not a self-map cannot be iterated inside \(X\) without extra work.
Completeness is different from compactness; Banach theorem does not require compactness.
Worked example
Example: Why does \(T(x)=x/2\) on \(X=(0,1)\) fail to have a fixed point in \(X\)?
It is a contraction because \(|T(x)-T(y)|=|x-y|/2\), and it maps \((0,1)\) into \((0,1)\). But the fixed point equation \(x=x/2\) gives \(x=0\), which is not in \((0,1)\). The missing endpoint shows why completeness matters.
Try it
Try it: A contraction on an incomplete metric space may fail to have:
Hint: The iteration can approach a limit that is not a point of the space.
Continuity on compact convex sets gives existence, not uniqueness
Learning goal: Use Brouwer-style hypotheses when contraction estimates are unavailable.
Key idea
In one dimension, every continuous map \(f:[a,b]\to[a,b]\) has a fixed point. Let \(g(x)=f(x)-x\). Since \(f(a)\ge a\) and \(f(b)\le b\), the intermediate value theorem gives some \(c\) with \(g(c)=0\), so \(f(c)=c\). In \(\mathbb{R}^n\), Brouwer theorem says a continuous self-map of a compact convex set has at least one fixed point.
Practical tests
Continuity: required for Brouwer-style existence.
Self-map: the set must map into itself.
Compact convex set: the standard finite-dimensional setting for Brouwer theorem.
Existence only: uniqueness needs stronger assumptions such as contraction.
No iteration promise: Brouwer alone does not say repeated iteration converges.
Worked example
Example: Find a fixed point of \(f(x)=1-x\) on \([0,1]\).
The map is continuous and sends \([0,1]\) to itself, so existence is guaranteed. Solving \(x=1-x\) gives \(x=1/2\). This map is not a contraction in the usual metric because \(|f(x)-f(y)|=|x-y|\).
Try it
Try it: A continuous map \([0,1]\to[0,1]\) is guaranteed to have:
Hint: Apply the one-dimensional fixed point argument to \(f(x)-x\).
Read the hypotheses before choosing Banach or Brouwer
Learning goal: Decide which fixed point principle fits the data of a problem.
Key idea
Banach theorem is metric and quantitative: it needs a contraction on a complete metric space and returns a unique fixed point plus convergence of iteration. Brouwer theorem is topological and finite-dimensional: it needs a continuous self-map of a compact convex set and returns at least one fixed point.
Hypotheses that matter
Use Banach when you can prove a global shrink factor \(k<1\).
Use Brouwer when you have continuity, compactness, convexity, and a self-map but no contraction.
Interval problems often reduce to the intermediate value theorem.
Uniqueness usually comes from a contraction or monotonicity argument, not from Brouwer alone.
Iteration convergence is a Banach conclusion, not a Brouwer conclusion.
Worked example
Example: Why does \(f(x)=\cos x\) have a fixed point in \([0,1]\)?
The map is continuous, and \(\cos x\in[\cos 1,1]\subset[0,1]\) for \(x\in[0,1]\), so the interval theorem gives a fixed point. In fact \(|f'(x)|=|\sin x|\le \sin 1<1\) on \([0,1]\), so Banach theorem also gives uniqueness and convergence of iteration.
Try it
Try it: A compact convex subset of \(\mathbb{R}^n\) with a continuous self-map is the setting of:
Hint: The theorem is about continuous self-maps of compact convex sets in finite-dimensional Euclidean space.
Most mistakes apply the right theorem with a missing hypothesis
Learning goal: Finish with distinctions that prevent common fixed point errors.
Common traps
Fixed point versus zero: solve \(T(x)=x\), not automatically \(T(x)=0\).
Fixed point versus contraction: a map can have a fixed point and still fail to be a contraction.
Lipschitz \(1\): nonexpansive is not the same as contraction.
Many fixed points: the identity map on \([0,1]\) fixes infinitely many points, so uniqueness needs a stronger argument.
Completeness: contractions can miss fixed points in incomplete spaces.
Self-map: check \(T(X)\subseteq X\) before invoking a theorem.
Brouwer: existence does not imply uniqueness.
Iteration: convergence of \(x_{n+1}=T(x_n)\) is guaranteed by Banach hypotheses, not by every fixed point theorem.
Worked example
Example: Explain why \(T(x)=2-x\) on \(\mathbb{R}\) has a fixed point but is not a contraction.
The fixed point equation \(x=2-x\) gives \(x=1\). However, \(|T(x)-T(y)|=|(2-x)-(2-y)|=|x-y|\), so the best Lipschitz constant is \(1\), not less than \(1\). Banach theorem does not apply.
Try it
Try it: Does Brouwer theorem by itself guarantee uniqueness of the fixed point?
Hint: A continuous self-map may have many fixed points, such as the identity map on a compact interval.
Final recap
A fixed point solves \(T(x)=x\).
A contraction has \(d(Tx,Ty)\le k d(x,y)\) for one \(k<1\).
Nonexpansive means Lipschitz constant at most \(1\); Banach needs strictly less than \(1\).
Constant maps shrink all distances to \(0\), while identity maps can have many fixed points.
Banach theorem needs a complete metric space and a contraction self-map.
Banach theorem gives existence, uniqueness, and convergence of Picard iteration.
The one-step error satisfies \(d(Tx,x^*)\le k\,d(x,x^*)\).
The uniqueness proof compares two fixed points and forces their distance to be \(0\).
Completeness keeps the Cauchy limit inside the space.
A continuous map \([a,b]\to[a,b]\) has a fixed point.
Brouwer theorem gives existence for continuous self-maps of compact convex subsets of \(\mathbb{R}^n\).
Brouwer theorem does not by itself give uniqueness or iteration convergence.
Next step: Close this lesson and try the quiz again. For each question, first ask whether it is testing the equation \(T(x)=x\), the contraction inequality, nonexpansive versus contraction, Banach hypotheses, Brouwer existence, or a missing-hypothesis trap.
Streak 5+
Streak 10+
Streak 15+
Streak 20+
Streak 25+