Professional Documents
Culture Documents
PDF Analysis On Polish Spaces and An Introduction To Optimal Transportation D J H Garling Ebook Full Chapter
PDF Analysis On Polish Spaces and An Introduction To Optimal Transportation D J H Garling Ebook Full Chapter
https://textbookfull.com/product/spaces-an-introduction-to-real-
analysis-tom-lindstrom/
https://textbookfull.com/product/functional-analysis-an-
introduction-to-metric-spaces-hilbert-spaces-and-banach-algebras-
second-edition-joseph-muscat/
https://textbookfull.com/product/transportation-engineering-an-
introduction-3rd-edition-c-jotin-khisty/
https://textbookfull.com/product/an-introduction-to-real-
analysis-first-edition-agarwal/
An Introduction to Classical Real Analysis Karl R.
Stromberg
https://textbookfull.com/product/an-introduction-to-classical-
real-analysis-karl-r-stromberg/
https://textbookfull.com/product/an-introduction-to-the-theory-
of-reproducing-kernel-hilbert-spaces-1st-edition-vern-i-paulsen/
https://textbookfull.com/product/an-introduction-to-the-theory-
of-reproducing-kernel-hilbert-spaces-1st-edition-vern-i-
paulsen-2/
https://textbookfull.com/product/an-introduction-to-performance-
analysis-of-sport-2nd-edition-adam-cullinane/
https://textbookfull.com/product/an-introduction-to-statistical-
methods-and-data-analysis-7th-edition-r-lyman-ott/
London Mathematical Society Student Texts 89
D. J. H. GARLING
Emeritus Reader in Mathematical Analysis, University of Cambridge,
and Fellow of St John’s College, Cambridge
University Printing House, Cambridge CB2 8BS, United Kingdom
One Liberty Plaza, 20th Floor, New York, NY 10006, USA
477 Williamstown Road, Port Melbourne, VIC 3207, Australia
314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre, New Delhi – 110025, India
79 Anson Road, #06–04/06, Singapore 079906
www.cambridge.org
Information on this title: www.cambridge.org/9781108421577
DOI: 10.1017/9781108377362
© D. J. H. Garling 2018
This publication is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 2018
Printed in the United Kingdom by Clays, St Ives plc
A catalogue record for this publication is available from the British Library.
Library of Congress Cataloging-in-Publication Data
Names: Garling, D. J. H., author.
Title: Analysis on Polish spaces and an introduction to optimal
transportation / D.J.H. Garling (University of Cambridge).
Other titles: London Mathematical Society student texts ; 89.
Description: Cambridge, United Kingdom ; New York, NY : Cambridge University
Press, 2018. | Series: London Mathematical Society student texts ; 89 |
Includes bibliographical references and index.
Identifiers: LCCN 2017028186 | ISBN 9781108421577 (hardback ; alk. paper) |
ISBN 1108421571 (hardback ; alk. paper) | ISBN 9781108431767 (pbk. ; alk. paper) |
ISBN 1108431763 (pbk. ; alk. paper)
Subjects: LCSH: Polish spaces (Mathematics) | Mathematical analysis. |
Transportation problems (Programming) | Topology.
Classification: LCC QA611.28 .G36 2018 | DDC 514/.32–dc23
LC record available at https://lccn.loc.gov/2017028186
ISBN 978-1-108-42157-7 Hardback
ISBN 978-1-108-43176-7 Paperback
Cambridge University Press has no responsibility for the persistence or accuracy
of URLs for external or third-party internet websites referred to in this publication
and does not guarantee that any content on such websites is, or will remain,
accurate or appropriate.
Contents
Introduction 1
2 Metric Spaces 18
2.1 Metric Spaces 18
2.2 The Topology of Metric Spaces 21
2.3 Completeness: Tietze’s Extension Theorem 24
2.4 More on Completeness 27
2.5 The Completion of a Metric Space 29
2.6 Topologically Complete Spaces 31
2.7 Baire’s Category Theorem 33
2.8 Lipschitz Functions 35
4 Semi-continuous Functions 50
4.1 The Effective Domain and Proper Functions 50
4.2 Semi-continuity 50
4.3 The Brézis–Browder Lemma 53
4.4 Ekeland’s Variational Principle 54
v
vi Contents
6 Càdlàg Functions 71
6.1 Càdlàg Functions 71
6.2 The Space (D[0, 1], d∞ ) 72
6.3 The Skorohod Topology 73
6.4 The Metric dB 75
7 Banach Spaces 79
7.1 Normed Spaces and Banach Spaces 79
7.2 The Space BL(X) of Bounded Lipschitz Functions 82
7.3 Introduction to Convexity 83
7.4 Convex Sets in a Normed Space 86
7.5 Linear Operators 88
7.6 Five Fundamental Theorems 91
7.7 The Petal Theorem and Daneš’s Drop Theorem 95
8 Hilbert Spaces 97
8.1 Inner-product Spaces 97
8.2 Hilbert Space; Nearest Points 101
8.3 Orthonormal Sequences; Gram–Schmidt Orthonormalization 104
8.4 Orthonormal Bases 107
8.5 The Fréchet–Riesz Representation Theorem; Adjoints 108
1
2 Introduction
Topological Properties
1
General Topology
This chapter contains a brief account of topological spaces and their properties.
It contains definitions and statements of fundamental results, and describes
the notation that is used. Proofs are generally not given; they can be found in
[G II] (and elsewhere).
(i) ∅ ∈ τ and X ∈ τ ;
(ii) if F is a finite subset of τ , then ∩U∈F U ∈ τ ; and
(iii) if G is any subset of τ , then ∪U∈G U ∈ τ .
(x − δ, x + δ) = {y : x − δ < y < x + δ} ⊆ U.
9
10 General Topology
Exercise 1.1.1 Suppose that X and Y are topological spaces, and that f is a
mapping from X into Y. Then f is continuous if and only if the graph mapping
G( f ) is a homeomorphism of X onto f .
Suppose that (xn )∞n=1 is a sequence in a topological space (X, τ ) and that
x ∈ X. Then xn converges to x if for each N ∈ Nx there exists n0 such that
xn ∈ N for each n ≥ n0 ; if so, we write that xn → x as n → ∞. x is an
accumulation point or limit point of the sequence if for each N ∈ Nx and each
n ∈ N there exists m ≥ n such that xm ∈ N. If xn → x as n → ∞, then x is a
limit point of the sequence.
Suppose that f is a continuous mapping from a topological space (X, τ )
into a topological space (Y, σ ), and that x ∈ X. f is sequentially continuous
at x if f (xn ) → f (x) as n → ∞ whenever xn → x as n → ∞, and is
sequentially continuous on X if it is sequentially continuous at each point of X.
A continuous mapping is sequentially continuous; as we shall see, the converse
is generally not true.
12 General Topology
There are conditions that control the size of a topological space, and of
topologies. A topological space is separable if there is a countable dense sub-
set. It is first countable if every point has a countable base of neighbourhoods,
and it is second countable if there is a countable base for the topology. These
notions are related in the following way.
Exercise 1.1.2 (i) A second countable space is first countable and
separable.
(ii) A subspace of a first countable space is first countable, and the product
of countably many first countable spaces is first countable.
(iii) A subspace of a second countable space is second countable, and the
product of countably many second countable spaces is second countable.
(iv) The product of countably many separable topological spaces is separable.
(But see Proposition 1.1.3.)
(v) A mapping from a first countable topological space into a topological
space is continuous if and only if it is sequentially continuous.
(vi) The topological space (R, τr ) is first countable and separable, but is not
second countable.
(Proofs can be found in [G II], Propositions 13.5.1 and 13.5.3.)
Products of separable spaces behave remarkably well; this illustrates the fact
that a product topology is a weak topology.
Proposition 1.1.3 Suppose that (X, τ ) is a separable topological space. Then
X (0,1] , with the product topology, is separable.
Proof Let C be a countable dense subset of X, and let Y be the space of ele-
ments of X (0,1] which take constant values in C on each of the intervals (i/k,
(i+1)/k] for 1 ≤ i ≤ k, for some k. Then Y is countable, and dense in X (0,1] .
There are also conditions which ensure that points and closed sets can be
distinguished topologically. A topological space (X, τ ) is
• a T1 space if singleton sets are closed, so that finite sets are closed;
• a T2 space, or Hausdorff space, if whenever x and y are distinct points of X
there exist disjoint open sets U and V with x ∈ U and y ∈ V;
• a T3 space if whenever A is a closed subset of X and x ∈ A there exist disjoint
open sets U and V with x ∈ U and A ⊆ V;
• a T4 space if whenever A and B are disjoint closed sets there exist disjoint
open sets U and V with A ⊆ U and B ⊆ V.
A topological space is a T3 space if and only if every point has a base of
neighbourhoods consisting of closed sets. A Hausdorff T3 space is called a
1.1 Topological Spaces 13
Let D be the set of dyadic fractions in [0, 1]. Arguing inductively, if r = p/2n ,
with p odd, define a closed set Ar such that
A(p−1)/2n ⊆ Aint
r ⊆ Ar ⊆ A(p+1)/2n ⊆ A(p+1)/2n ;
int
(p − 1 and p + 1 are even, so that A(p−1)/2n and A(p+1)/2n have already been
defined). Now if x ∈ X let f (x) = inf{r ∈ D : x ∈ Ar } (where inf(∅) = 1).
Then f (x) = 0 for x ∈ A, f (x) = 1 for x ∈ B and 0 ≤ f (x) ≤ 1. It remains
to show that f is continuous. But if 0 < s ≤ 1 then f (x) < s if and only if
x ∈ Cs = ∪s<r≤1 (X \ Ar ), and if 0 ≤ t < 1 then f (x) > t if and only if
x ∈ Dt = ∪0≤r≤t Aint
r . Then Cs and Dt are open sets, from which it follows that
f is continuous.
Consequently, a normal space is completely regular.
Theorem 1.1.9 A topological space (X, τ ) is completely regular if and only if
it is homeomorphic to a subspace of a hypercube.
Proof The condition is certainly sufficient. Suppose that (X, τ ) is completely
regular. Let
A = {(x, F) : x ∈ X, F closed in X, x ∈ F}.
If α = (x, F) ∈ A there exists a continuous function fα : X → [0, 1] such that
fα (x) = 0 and f (y) = 1 for y ∈ F. Let f (x) = {fα (x) : α ∈ A}. If x = y then
β = (x, {y}) ∈ A, and fβ (x) = 0 = 1 = fβ (y), and so f is injective. Since each
fα is continuous, f : X → [0, 1]A is continuous. Conversely if x ∈ X and U is
an open neighbourhood of x then γ = (x, X \U) ∈ A, V = {g ∈ f (X) : gγ < 1}
is a neighbourhood of f (x) in f (X), and V ⊆ f (U), so that f −1 : f (X) → X is
also continuous.
Corollary 1.1.10 A second countable normal space is homeomorphic to a
subspace of the Hilbert cube H.
Proof Replace A by the countable set of pairs {(xn , Cm )}, where (Cm )∞
m=1
is a countable base of closed sets and (xn )∞
n=1 is a dense sequence, and
xn ∈ Cm .
Note that the theorem uses the axiom of choice, but the corollary does not.
1.2 Compactness 15
Let us end this section by observing that the definition of a topological space
that we have given is short and easy, but is a little misleading. The important
fact is that topology is a local phenomenon; it is therefore often appropriate
to define a topology in terms of neighbourhoods. The neighbourhoods Nx of a
point x form a filter; that is,
(i) ∅ ∈ Nx ;
(ii) if N ∈ Nx and N ⊆ M then M ∈ Nx ;
(iii) if N1 , N2 ∈ Nx then N1 ∩ N2 ∈ Nx .
Exercise 1.1.11 Suppose that (X, τ ) is a topological space and that {Nx }x∈X is
the set of neighbourhood filters. Show that if N ∈ Nx then x ∈ N and there
exists O ∈ Nx such that O ⊆ N and O ∈ Ny for each y ∈ O.
Conversely, if {Nx }x∈X is a set of filters on a set X which has these properties,
show that this defines a topology on X for which {Nx }x∈X is the set of
neighbourhood filters.
1.2 Compactness
Suppose that A is a subset of a topological space (X, τ ). A collection O of
open sets is an open cover of A if A ⊆ ∪O∈O O. An open cover is finite if it
has finitely many members. The set A is compact if every open cover of A has
a finite subcover. Compact sets are a topological approximation to finite sets.
By considering complements, it follows that (X, τ ) is compact if and only if it
has the finite intersection property; if A is a collection of closed sets with the
property that ∩A∈Af A is not empty for each finite collection Af of sets in A,
then the total intersection ∩A∈A A is not empty.
Here are some basic properties of compact sets.
Exercise 1.2.1 (i) The union of a finite set of compact sets is compact.
(ii) A compact Hausdorff space is normal.
(iii) A subset of a compact Hausdorff space is compact if and only if it is
closed.
(iv) The continuous image of a compact set is compact.
(v) A continuous bijection of a compact space onto a Hausdorff space is a
homeomorphism.
For details, see [G II], Propositions 15.1.3, 15.1.4, Corollary 15.1.7.
Here is one of the fundamental results of general topology. It requires the
axiom of choice, and indeed is equivalent to it; Tychonoff’s theorem is true if
and only if the axiom of choice holds.
16 General Topology
In this chapter, we introduce the idea of a metric space. Metric spaces have a
natural topology, which we describe in the first two sections of the chapter, but
they have more structure than that. They also have a uniform structure, which
we shall consider in detail in Chapter 5, and which is the setting for the notion
of completeness, which we discuss in Sections 2.3 to 2.7, a Lipschitz structure,
which we consider in Section 2.8, and a geometric structure.
18
2.1 Metric Spaces 19
(i) The function d(x, y) = |x−y| is the usual metric on R and the usual metric
1
on C. Similarly the function d(x, y) = ( nj=1 |xj − yj |2 ) 2 is the usual
metric, or Euclidean metric, on Rn and the usual metric, or Hermitian
metric, on Cn .
(ii) If X is any set, the function defined as d(x, y) = 1 if x = y and d(x, y) = 0
if x = y is the discrete metric on X.
(iii) If Y is a subset of a metric space (X, d) then the restriction of d to Y × Y
is a metric on Y, the subspace metric.
(iv) If A is a non-empty subset of a metric space, the diameter diam(A) is
defined to be sup{d(a, b) : a, b ∈ A}. Note that it follows from the
quadrilateral inequality that diam(A) = diam(A). A set A is bounded if A
is empty or diam(A) < ∞.
Suppose that S is a non-empty set, that (Y, ρ) is a metric space and that
f is a mapping from S to Y. We define the oscillation ωS (f ) of f on S to be
diam(f (S)) = sup{ρ(f (s), f (t)) : s, t ∈ S}. f is bounded on S if ωS (f ) < ∞,
and B(S, Y) denotes the set of bounded mappings from S to Y. The function
d∞ (f , g) = sups∈S ρ(f (s), g(s)) is then a metric on B(S, Y), the uniform metric
on B(S, Y). A sequence (fn )∞ n=1 converges to f in this metric if and only if
fn (s) → f (s) uniformly in s; sups∈S |fn (s) − f (s)| → 0 as n → ∞; thus
convergence in d∞ is called uniform convergence. We denote B(S, R) by
B(S), and we denote (B(N, Y) by l∞ (Y); thus l∞ (Y) is the space of bounded
sequences in Y.
Another random document with
no related content on Scribd:
The Project Gutenberg eBook of Kuin unikuva
This ebook is for the use of anyone anywhere in the United States
and most other parts of the world at no cost and with almost no
restrictions whatsoever. You may copy it, give it away or re-use it
under the terms of the Project Gutenberg License included with this
ebook or online at www.gutenberg.org. If you are not located in the
United States, you will have to check the laws of the country where
you are located before using this eBook.
Language: Finnish
Kirj.
Eliza Orzeszko
Suomennos
Oi! hän ei totta tosiaan olisi toivonut, että niin olisi käynyt, mutta ei
auttanut. Halvinkin hame vaati rahaa, ja hänen täytyi kovasti
kamppailla, jotta isän palkka vaan kaikkeen riittäisi. Tähän saakka ei
toki mitään ollut puuttunut, joskin isä sai kieltäymyksiä kestää, sillä
heikko kun terveys oli, olisi ravitsevampi ruoka ollut tarpeellinen,
varsinkin hedelmät…
Ensin hän kulki pää alas painuneena, mutta sitte kohosi katse ja
ihaili puiston puita. Ne seisoivat liikkumattomina hiljaisessa ilmassa,
valossa syysauringon, jonka kultaiset säteet siellä täällä murtautuivat
läpi kellastuneiden tai punettuneiden lehtien. Joskus kahisi kuihtunut
lehti kävelevän jalkojen alla, joka yhä hiljensi askeleitaan, antaen
katseensa liukua puiden latvoista, jotka punakeltaisina hohtivat,
pitkin paksuja runkoja, vihreitten köynnösten kietomia.
Hän teki sen johtopäätöksen, että tämä puisto oli viehättävä, joskin
vain varsin pieni, pikkukaupungin puisto. Mutta ehkäpä se oli
viehättävä juuri siksi, että täällä vallitsi suurkaupungissa mahdoton
hiljaisuus.
Tosin nuo eivät olleet varsin viisaita unelmia, mutta täällä, tässä
ympäristössä, ne vastustamattomat mielikuvitukset kohosivat,
jättäen joksikin aikaa kaihomieltä sydämen sisimpään. Ja mitä sitte
on tässä maailmassa viisasta?
— Me siellä asumme.
— Viisitoista vuotta.
He vaikenivat. Uudelleen hämmennyksissään painoi tyttö päänsä
työhön ja alkoi taas ommella; mies nojasi aitaa vastaan, häntä
katsellen, ollenkaan yrittämättä lähteä. Juuri tuo katseleminen pani
tytön hämilleen.
— Aivan niin.
Mutta näytti siltä kuin lukija samalla olisi jotakin miettinyt, sillä
yht'äkkiä hän sulki kirjan ja kumartaen lausui:
— Tietysti. Kun on niin rikas, tottakai voi tehdä mitä ikinä tahtoo.
Hän nauroi.