logo

Lefschetz Fixed Point Theorem Proof 📂Banach Space

Lefschetz Fixed Point Theorem Proof

Theorem 1

Let’s denote the scalar field of norm space (X,)(X , \left\| \cdot \right\|) as C\mathbb{C}. Then

XX is finite-dimensional.     \iff B(0;1)\overline{ B ( 0 ; 1 ) } is compact.

Explanation

B(0;1):={xX:x1}\overline{ B ( 0 ; 1 ) } := \left\{ x \in X : \| x \| \le 1 \right\} denotes the closed unit ball. According to the Riesz’s theorem, to determine whether the entire space is finite-dimensional, it is enough to check a very small region. It is often not considered necessary and sufficient conditions for finite-dimensional normed spaces, which makes this theorem very mathematical.

Proof

Strategy: A simple homeomorphism from the manageable Cn\mathbb{C}^{n} to XX is given, transferring the compactness from Cn\mathbb{C}^{n} to XX. In reverse, based on the compactness of B(0;1)\overline{ B ( 0 ; 1 ) }, a finite-dimensional vector space is created, and it is shown that it actually includes XX.


  • (    )(\implies)

    If we consider dimX=n\dim X = n, there exists a basis {e1,,en}\left\{ e_{1} , \cdots , e_{n} \right\} for XX. If we define the function f:(Cn,1)(X,)f : ( \mathbb{C}^{n} , \| \cdot \|_{1} ) \to (X , \| \cdot \| ) as f(λ1,,λn):=λ1e1++λnenf(\lambda_{1} , \cdots , \lambda_{n} ) : = \lambda_{1} e_{1} + \cdots + \lambda_{n} e_{n}, ff is a continuous bijection.

    The closed unit ball B1(0;1)={(λ1,,λn)Cn  λ1++λn1}\overline{ B_{ \| \cdot \|_{1} } ( 0 ; 1 ) } = \left\{ (\lambda_{1} , \cdots , \lambda_{n}) \in \mathbb{C}^{n} \ | \ | \lambda_{1} | + \cdots + | \lambda_{n} | \le 1 \right\} in Cn\mathbb{C}^{n} is compact by the Heine-Borel theorem. Since ff is continuous, f(B1(0;1))f \left( \overline{ B_{ \| \cdot \|_{1} } ( 0 ; 1 ) } \right) is also compact.

    Meanwhile, since λ1e1++λnenλ1++λn\| \lambda_{1} e_{1} + \cdots + \lambda_{n} e_{n} \| \le | \lambda_{1} | + \cdots + | \lambda_{n} |, it follows B(0;1)f(B1(0;1))\overline{ B ( 0 ; 1 ) } \subset f \left( \overline{ B_{ \| \cdot \|_{1} } ( 0 ; 1 ) } \right). Since B(0;1)\overline{ B ( 0 ; 1 ) } is a closed subset of the compact set f(B1(0;1))f \left( \overline{ B_{ \| \cdot \|_{1} } ( 0 ; 1 ) } \right), it is compact.

  • (    )(\impliedby)

    Let’s consider 0<ε<10 < \varepsilon < 1.

    Since B(0;1)\overline{ B ( 0 ; 1 ) } is compact, there exists a finite subcover that satisfies B(0;1)i=1mB(xi;ε)\displaystyle \overline{ B ( 0 ; 1 ) } \subset \bigcup_{i=1}^{m} B \left( x_{i} ; \varepsilon \right) for the open cover xB(0;1)B(x;ε)\displaystyle \bigcup_{x \in \overline{ B ( 0 ; 1 ) } } { B \left( x ; \varepsilon \right) }. Let’s say M:=span{x1,,xn}M := \text{span} \left\{ x_{1} , \cdots , x_{n} \right\}.

    B(0;1)i=1mB(xi;ε)\displaystyle \overline{ B ( 0 ; 1 ) } \subset \bigcup_{i=1}^{m} B \left( x_{i} ; \varepsilon \right) means that B(0;1)mMB(m;ε)\displaystyle \overline{ B ( 0 ; 1 ) } \subset \bigcup_{ m \in M } B \left( m ; \varepsilon \right) holds. Since mspan{x1,,xn} m \in \text{span} \left\{ x_{1} , \cdots , x_{n} \right\} from the beginning, no matter how small the diameter ε\varepsilon is chosen, this inclusion relation continues to hold. Therefore, for kNk \in \mathbb{N}

    B(0;1)mMB(m;ε)mMB(m;ε2)mMB(m;εk) \overline{ B ( 0 ; 1 ) } \subset \bigcup_{ m \in M } B \left( m ; \varepsilon \right) \subset \bigcup_{ m \in M } B \left( m ; \varepsilon^2 \right) \subset \cdots \subset \bigcup_{ m \in M } B \left( m ; \varepsilon^k \right)

    Now, considering any non-zero vector xXx \in X, for some ykMy_{k} \in M, zk:=B(0;εk)\displaystyle z_{k} : = B ( 0 ; \varepsilon^k )

    xx=yk+zk {{ x } \over { \| x \| }} = y_{k} + z_{k}

    When kk \to \infty, since zk0z_{k} \to 0

    yk=xxzkxxM=M y_{k} = {{ x } \over { \| x \| }} - z_{k} \to {{ x } \over { \| x \| }} \in \overline{ M } = M

    That is, since xMx \in M, and XMX \subset M, but since MXM \subset X

    X=span{x1,,xn} X = \text{span} \left\{ x_{1} , \cdots , x_{n} \right\}

    Therefore, XX is a finite-dimensional vector space.

See Also

Generalization from Euclidean Spaces

Riesz’s theorem points out the compactness of the closed unit ball B(0;1)\overline{B (0;1)} in norm spaces as an equivalent condition of finiteness of dimensions. Since the kk-cell [0,1]k[0,1]^{k} is compact in Euclidean spaces, and there exists a homeomorphism with the closed unit ball, Riesz’s theorem can be seen as a generalization concerning the compactness of the kk-cell.


  1. Kreyszig. (1989). Introductory Functional Analysis with Applications: p80. ↩︎