Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Closedness and Lower Semicontinuity

  1. Jan 11, 2006 #1
    Given a real-valued function [tex]f:X \mapsto \mathbb{R}[/tex] where [tex]X[/tex] is a subset of [tex]\mathbb{R}^n[/tex]. Consider a closed extended real-valued function [tex]\tilde{f}:\mathbb{R}^n \mapsto (-\infty, \infty][/tex] defined as

    \tilde{f}(x) = \left \{
    f(x) & \quad , \textup{ if } x \in X \\
    \infty & \quad , \textup{ otherwise}

    I want to show that if [tex]X[/tex] is not closed, then for a sequence [tex]\{x_k\} \subset X[/tex] converging to a point [tex]\bar{x}[/tex] on the boundary of [tex]X[/tex] that does not belong to [tex]X[/tex] necessarily results in [tex]\liminf_{k \to \infty} f(x_k) = \infty[/tex]. I use the following definition for closed function (definition used by my text book): A function is said to be closed if its epigraph is closed.

    The following is my proof, but I am a bit unsure of its correctness. Hope someone can give me some feedback.

    Suppose otherwise that [tex]\liminf_{k \to \infty} f(x_k) = M[/tex] for some [tex]M < \infty[/tex]. Since [tex]\tilde{f}[/tex] is closed, it is lower semicontinuous. Hence,

    \tilde{f}(\bar{x}) \leq \liminf_{k \to \infty} \tilde{f}(x_k) = \liminf_{k \to \infty} f(x_k) = M < \infty.

    But [tex]\tilde{f}(\bar{x}) = \infty[/tex], thus a contradiction.
    Last edited: Jan 11, 2006
  2. jcsd
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Can you offer guidance or do you also need help?
Draft saved Draft deleted

Similar Discussions: Closedness and Lower Semicontinuity
  1. Lower integrals (Replies: 4)

  2. Lowe Semicontinuity (Replies: 10)

  3. Upper and Lower Sums (Replies: 5)