• 検索結果がありません。

five lemma

N/A
N/A
Protected

Academic year: 2022

シェア "five lemma"

Copied!
4
0
0

読み込み中.... (全文を見る)

全文

(1)

II

:

2006

11

2

(

)

1. five lemma

.

! #"%$'&)(+*,)-.0/2143

.

−−−→ A −−−→ f 1 B −−−→ f 2 C −−−→ f 3 D −−−→ f 4 E −−−→

α

y β

y γ

y δ

y ε

 y

−−−→ A 0 −−−→

f 1 0 B 0 −−−→

f 2 0 C 0 −−−→

f 3 0 D 0 −−−→

f 4 0 E 0 −−−→

5768+9%:<;>=4?@

3A'B

,

C

68+9%:

α, β, δ, ε

:'DFE7G+HI?@

3AKJ#3

.

L)MA0N

, γ

O

D)EP?4@

34LIA<QRS

.

2. f

C

T2UWVWX

N

YZ+[]\^[]_I`<aW\

D0bWG2H

A7B

, N × [0, 1]

9

, (x, 0) ∼ (f(x), 1)

?0c+d.eQf

3

D)ghFi

kj

f

, M = N × [0, 1]/ ∼

Aml%nIJ#3

. M

:

C

T7U7V7X

?I@po

, mapping torus

A!q2r

f 3

. (1)

L<SAN

0 → H k− 1 (N, R )

Image(id −f ) → H k (M, R ) → H k (N, R ) f → 0

A!sut

;%=Wv+w

kxSy

.

zW{B

f

:

, f

)|7x4JR3~}

D)E

f : H k (N, R ) → H k (N, R )

?

@po

, H k− 1 (N, R ) f

:

, f

?2

l

ef

3~€W\7‚AKJ3

. (2) N = S 1 , f (e ) = e −iθ

A7Bkƒ

,

„)…

d9

o

, M

~l+n4J3

. M

:

Klein

?I@

34LIA0!‡ˆBkƒ

,

‰7IŠ0‹+ŒP7Ž! !R

.

‘%’

.

2

?I:

,

“2”•BFz'–0—

?™˜

–

ef

ƒ2s2z€2\QšpB#›œRJ34LIA

.

‰+€W\

9

BkY>ž

:FŸ

1 Z f4¡

s

.

(2)

Š<‹WŒP%Ž!

‘, R ’

–%J34L4A J#3

.

1.

•B

9

™JR3

. (1) α

=

, β

A

δ

P?I@

3 ¡

ZQr

, γ

:

P?4@

3

. c ∈ Ker γ

A J3

.

1. δf 3 (c) = f 3 0 γ (c) = 0

?4@o

, δ

:

{>Y•Z

, f 3 (c) = 0

?4@

3

. C

9

yI3

;%=

Y

Z

c = f 2 (b)

A

¡

3I#t

¡

b ∈ B

4J3

. 2. f 2 0 β(b) = γf 2 (b) = γ (c) = 0

?4@

3

. B 0

9

y23

;7=

YpZ

, β(b) = f 1 0 (a 0 )

A

¡

34

t ¡

a 0 ∈ A 0

IJ3

. α

=

{>YpZ

, α(a) = a 0

A

¡

3#t

¡

a ∈ A

IJ3

. 3. βf 1 (a) = f 1 0 α(a) = f 1 0 (a 0 ) = β(b)

?I@

3

. β

S9

o

, f 1 (a) = b

?4@

3

. 4. c = f 2 (b) = f 2 f 1 (a)

?P@

3

,

;7=

(

9+:Wd

0

9%¡

3PLPA

?

\

)

9 o

, c = 0

?I@

3

.

, γ

P?I@

3ILIA‡

eFf

z

. (2) ε

, β

A

δ

=P?4@

3 ¡

Z~r

, γ

:F=P?I@

3

. c 0 ∈ C 0

A J3

.

1. f 3 0 (c 0 )

</W143RA

, δ

=P?I@

34LA YpZ

, δ(d) = f 3 0 (c 0 )

A

¡ 3

d ∈ D

IJ3

. 2. εf 4 (d) = f 4 0 δ(d) = f 4 0 f 3 0 (c 0 )

?

,

L

f#:

0

?P@

3

. ε

:

#?4@

3YZ

, f 4 (d) = 0

?I@

3

. D

9

yW3

;%=

YpZ

f 3 (c) = d

A

¡ 3

c

IJ3

.

3. f 3 0 γ(c) = δf 3 (c) = δ(d) = f 3 0 (c 0 )

?I@

30YpZ

, f 3 0 (c 0 − γ(c)) = 0

?I@

3

. C 0

9

y3

;

=M9

o

, c 0 − γ(c) = f 2 0 (b 0 )

A

¡ 3

b 0 ∈ B 0

IJR3

. 4. β

=?I@

30YZ

, b 0 = β(b)

A

¡ 3

b ∈ B

PJ3

. γf 2 (b) = f 0 2 β(b) = f 0 2 (b 0 ) = c 0 − γ(c)

?I@

3

.

BFz <ƒ

, c 0 = γ(f 2 (b) + c)

A

¡o

,

γ

:F=4?I@

3

.

!

6"

,

#%$™

ef

z

.

2. (1)

&('

“2”pBFz'–—

:)*

,+

o

:

- ‚

[0, 1]

I + = (0, 1), I − = [0, 1] \ { 1 2 }

A]\My

,

.0/ˆBkƒ

M = M + ∪ M −

A]\My3

. M + 3 [(x, t)] 7→ (x, t) ∈ N × I +

9 o

, M +

:

N × I +

A a+\

D<b4?@po

, M −

:

M − 3 [(x, t)] 7→

( (x, t) 0 ≤ t < 1 2

SAN

(f 1 (x), t − 1) 1 2 < t ≤ 1

SAN

∈ N × (− 1 2 , 1

2 )

9 o

, N × (− 1 2 , 1 2 )

A a%\

DFb?2@

3

.

z'{B

[(x, 0)] = [(f (x), 1)]

21N43(

,

„+25

I?

O * ,5

?

O

(x, 0)

A

¡

well-defined

?W@

3IL2A 9,67

B>4t

. N × I + , N × (− 1 2 , 1 2 )

:

,

8

9

N

9

‹+Œ:9<;(=>

?I@

3

.

B!Y.B4?@.

-,A

JR3'z7

9 &B

N + ,

C

B

N −

?

D("

J.LA

9

J3

.

(3)

, I + ∩ I − = (0, 1 2 ) t ( 1 2 , 1)

A 3

,

R \

I 0 ,

C

I 1

J

.

L0

A™N

M + ∩ M −

:

,

aW\

Db

M + ∼ = N × I +

+~ƒ

, N × I 0 t N × I 1

A]aW\

D0bP?I@

3

. N × I 0 , N × I 1

:

,

8

9

N

A ‹7Œ 9<;0= >

?@

3

.

B!YB4?@

-2A

J3>z%

9 &B

N 0 ,

C B

N 1

?D("

JLA

9

J3

. Mayer-Vietoris

;'=Ww

o

H k (M ) - H k (N + ) ⊕ H k (N − ) ϕ

- H k (N 0 ) ⊕ H k (N 1 ) H k− 1 (M ) - H k− 1 (N + ) ⊕ H k− 1 (N − )

ϕ

- H k− 1 (N 0 ) ⊕ H k− 1 (N 1 )

d

S3

.

L<SAN

ϕ

1

wD

‡4J3A

ϕ =

id id id f

A ¡ 3

. (

š#–

)

Ker ϕ = {u ⊕ v ∈ H k (N + ) ⊕ H k (N − ) | u + v = 0, u + f (v ) = 0} ∼ = H k (N ) f , Coker ϕ = H k (N 0 ) ⊕ H k (N 1 )

{x ⊕ y ∈ H k (N 0 ) ⊕ H k (N 1 ) | x = u + v, y = u + f (v)} ∼ = H k (N ) Image(id −f )

S3

.

0ƒ S3

.

——————————

–—

" o

——————————

ϕ

!œ4FJ43

.

–)—I

?'?

ƒ#NFzQa'\

Db

M + → N + ×I + , M − → N − ×(− 1 2 , 1 2 )

,

‰

f™f

F + , F −

?D("

J

.

(4)

ϕ

M ±

)ƒ N>JpA

,

H k (M + ) ⊕ H k (M − ) −−−→ Φ H k (M + ∩ M − )

F + ⊕F −

x

∼ = ∼ =

x

 ( F + | M+ M− ) H k (N + × I + ) ⊕ H k (N − × (− 1 2 , 1 2 )) H k (N 0 × (I 0 t I 1 ))

π + ⊕π

x

∼ = ∼ =

 y s

∗ 0 ⊕s 1

H k (N + ) ⊕ H k− 1 (N − ) −−−→

ϕ H k (N 0 ) ⊕ H k (N 1 )

A ¡ 3

.

„2

5

Φ

:

[α + ] ⊕ [α − ] 7→ h

α − | M + ∩M − α + | M + ∩M i

?4@•o

, π + : N + × I + → N + , π : N × (− 1 2 , 1 2 ) → N

:

d

\..3

G2H#?I@o

, s 0 , s 1

:

, N 0 → N 0 × (I 0 t I 1 )

?

,

‰

f ™f

x 7→ (x, 1 4 ), x 7→ (x, 3 4 )

?>Ÿ

1uZ

f

3PO<

?4@

3

. (Poincar´e

9™?

ƒNz

D<E+GWH

)l7n #s2{+‰t

.)

9 ˆBkƒ2sˆ›™A

ϕ([β + ] ⊕ [β − ])

= (s 0 ⊕ s 1 ) ◦

F + | M + ∩M −

− 1 ∗

◦ Φ ◦ (F + ⊕ F ) ◦ (π + ⊕ π )([β + ] ⊕ [β − ])

= (s 0 ⊕ s 1 ) ◦

F + | M + ∩M −

− 1 ∗

◦ Φ[(π + ◦ F + ) β + ] ⊕ [(π − ◦ F − ) β − ]

= (s 0 ⊕ s 1 ) ◦

F + | M + ∩M −

− 1 ∗ h

(π − ◦ F − ) β − | M + ∩M − (π + ◦ F + ) β + | M + ∩M i

= (s 0 ⊕ s 1 ) ◦

F + | M + ∩M −

− 1 ∗ h

( π − ◦ F − | M + ∩M ) β − − ( π + ◦ F + | M + ∩M ) β +

i

= h

( π − ◦ F − | M + ∩M ◦ F + | M + ∩M −

− 1

◦ s 0 ) β − − (π + ◦ F + | M + ∩M ◦ F + | M + ∩M −

− 1

◦ s 0 ) β +

i

⊕ h

( π − ◦ F − | M + ∩M ◦ F + | M + ∩M −

− 1

◦ s 1 ) β − − ( π + ◦ F + | M + ∩M ◦ F + | M + ∩M −

− 1

◦ s 1 ) β +

i

A ¡ 3

.

+d7GWH

~l+n

9

BFz )ƒ ˆBkƒ43A

, π − ◦ F − | M + ∩M ◦ F + | M + ∩M −

− 1

◦ s 1 = f 1

?

,

‰7O<

:

id

?I@

3

.

BFz <ƒ

ϕ([β + ] ⊕ [β − ]) = ([β − ] − [β + ]) ⊕ (f 1 ) [β − ] − [β + ]

A ¡ 3

.

L< 1

wD

‡4J

f r

,

− id id

− id (f 1 )

?I@

3

,

9

YZ 1 w

"!Sy

f r

,

„2#O<

9%¡

3

.

(2)

–

参照

関連したドキュメント

It is not difficult to see that constant-depth circuits with MOD p gates only (p prime), cannot compute even simple functions: the fan-in n OR or AND functions, since they can

We prove the existence of many complete graphs in almost all sufficiently dense partitions obtained by an application of Szemer´edi’s Regularity Lemma.. More precisely, we consider

Mocanu, On some classes of first order differential subordinations and univalent functions, Michigan Math.. Mocanu, Differential Subordinations: Theory and Ap- plications, Series

There are obvious relative and parametrised versions similar to those for the flattening lemma (and proved using those versions), and furthermore we can assume that ε is given by

In the same way as the classical Sperner’ Lemma corresponds to a homology theory of simplicial complexes (see, e.g. [2]), our infinitary version of Sperner’s Lemma (Theorem

In particular, in the case of a topological groupoid with open range map and a G-uniformity given by a fundamental system of neighborhoods of the unit space, our construction allows

著者 HINOHARA Yukitoshi, KAWASAKI Ken‑ichiroh,

direct product of compact Hausdorff spaces appears so as to the weaker version of Tychonoff's theorem will lead the result.. lemma is proved using the Ultrafilter Theorem