現実的なインフレーション模型は何かppp.ws/PPP2016/slides/terada.pdfUniversality...

現実的なインフレーション模型は何か

寺田 隆広Korean Institute of Advanced Study

Thoughts on realistic inflation models 2016

素粒子物理学の進展 2016, 基礎物理学研究所, 2016/9/6

DISCLAIMER

• 主にインフレーションのレビューです。

• 限られた経験/知識に基づき、偏見に満ちています。

• 個々の模型では、議論に色々な抜け道があります。

• 皆様の模型が出なくても怒らないでください。

　　　　　　　　　　　　　（コメントは歓迎。）

• 寄り道して関連した自分の仕事を紹介します。

Outline1. Introduction: inflation in a nutshell

2. Universality classes of inflationRealization by “pole inflation”

3. Initial conditions: Small-field or Large-field?

4. Shift symmetry and its originU(1): pNGB or Wilson line Weak Gravity Conjecture R: scale invariant models

5. Summary & Conclusion

3

Introduction: inflation in a nutshell

5

動機・利点• 指数関数的膨張によって、一様性問題、平坦性問題、モノポール問題を解決する。• インフラトンの量子揺らぎにより、宇宙の大規模構造の「種」をつくる。

宇宙の加速膨張

Rµ⌫ � 1

2gµ⌫R = 8⇡GTµ⌫

Tµ⌫ = diag(�⇢, P, P, P )

Einstein eq. 一様等方宇宙FLRW universe

Friedmann eq.

ds2 = �dt2 + a(t)2(dx2 + dy2 + dz2)

✓a

a

◆2

=8⇡G⇢

3

a

a= �4⇡G

3(⇢+ 3P )

Slow-roll

�+ 3H�+ V 0 = 0

slow-roll 近似

✏ =1

2

✓V 0

V

◆2

⌧ 1

|⌘| =����V 00

V

���� ⌧ 1

⇢ =1

2�2 + V

P =1

2�2 � V

a(t) ' eHt

P ' �⇢ ' const.

N =

Z tend

tHdt =

Z �

�end

d�p2✏

Minkowski 時空 Black Hole de Sitter 宇宙

加速する観測者Rindler horizon Event horizon

解析接続された世界

ホワイトホール

ブラックホール

Cosmological horizon

観測者r = 0

r = 1r = 0t = 1

観測者

Unruh radiation Hawking radiation Gibbons-Hawking radiationT =a

2⇡T =

2⇡T =

H

2⇡

inflaton fluctuation �� =H

2⇡

curvature perturbation ⇣ = �N = H�t =H

���

Power spectra

Pt(k) = At

✓k

k⇤

◆nt

Ps(k) = As

✓k

k⇤

◆ns�1

ns � 1 = �6✏+ 2⌘

r ⌘ At

As= 16✏

6

[Unruh, PRD14 (1976) 870] [Hawking, Commun.Math.Phys. 43 (1975) 199, Erratum: ibid. 46 (1976) 206] [Gibbons, Hawking, PRD15 (1977) 2738]

Excellent fit by ΛCDM

7

Scale invariant (ns ⇠ 1)

Adiabatic

Gaussian

(�iso

⇠ 0)

(fNL ⇠ 0)

ns = 0.9655± 0.0062

�iso

(0.002Mpc�1) < 4.1⇥ 10�2

(Planck TT+ low P)

(for CDM)

(Planck TT+ low P)f local

NL

= 0.8± 5.0

[Planck collaboration, 1502.02114, 1502.01592]

ns

r

Universality classes of inflation

Analogy to Renormalization Group

dg

d lnµ= �(g)

d�

d ln a= �(�)

The Hamilton-Jacobi formalism �(t) $ t(�)

“superpotential”→ This implies:

cf.)

W (�) ⌘ �H

W� = �/2 V = 3W 2 � 2W 2�

where �(�) = �2W�

W=

�

H= ±

s3(P + ⇢)

⇢= ±

p2✏

→ classified by the behavior near the fixed point (de Sitter).

9

Underlying connections?[McFadden, Skenderis, 0907.5542, 1001.2007]

See also, dS/CFT and FRW/CFT.[Strominger, hep-th/0106113][Witten hep-th/0106109][Larsen et al., hep-th/0202127][Halyo, hep-th/0203235]

[Freivogel et al., hep-th/0606204][Sekino et al., 0908.3844]

10

11

Universality classes of inflation[Garcia-Bellido et al., 1402.2059][Mukhanov, 1303.3925] [Roest, 1309.1285] [Binetruy et al., 1407.0820]

Figures from [Garcia-Bellido, Roest, 1402.2059]

Universality classes of inflation

12

13

Universality classes of inflation[Garcia-Bellido et al., 1402.2059][Mukhanov, 1303.3925] [Roest, 1309.1285] [Binetruy et al., 1407.0820]

観測的ステータス

◎◯△×

△△

Any underlying mechanism for the universality?

Inflationary Attractor ModelsModel space observables

ns

rpredictions

a limit of a parameter

ns

r“attraction”

14

Unity of cosmological attractors

15

[Galante, Kallosh, Linde, Roest, 1412.3797]

(p�g)�1L = �1

2⌦(�)R� 1

2KJ(�)(@µ�)

2 � VJ(�)

Unity of cosmological attractors

16

[Galante, Kallosh, Linde, Roest, 1412.3797]

2nd order pole(s) in ! KE

Inflation occurs near the pole.

Canonical normalization makes the potential flat.

KE(�) '3↵/2

(�� �0)2

(p�g)�1L = �1

2R� 1

2KE(')(@µ')

2 � VE(')

Pole inflation

17

[Galante, Kallosh, Linde, Roest, 1412.3797]

ns = 1� p

(p� 1)Nr =

8

ap

✓ap

(p� 1)N

◆ pp�1

[Broy, Galante, Roest, Westphal, 1507.02277]

�p�g��1 L = � ap

2'p@µ'@µ'� V0

�1� '+O('2)

�

V =

8<

:V0

✓1�

⇣p�22pap�⌘� 2

p�2+ · · ·

◆(p 6= 2),

V0

�1� e��/

pap + · · ·

�(p = 2),

Change of potential shape

18

The original potential

-0.8 -0.6 -0.4 -0.2

0.5

1.0

1.5

2.0

2.5

3.0

3.5

2 3 4 5 6

0.5

1.0

1.5

2.0

2.5

3.0

3.5

-1.0 -0.5 0.5 1.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

0 < p < 2 p � 2

“hilltop” “inverse-hilltop”

Inflation with a singular potential

19

The original diverging potential

“power-law” “chaotic”

-1.0 -0.5 0.0 0.5 1.0

10

20

30

40

0.5 1.0 1.5

20

40

60

80

100

120

140

2 3 4 5 6

10

20

30

40

50

p = 2 p > 2

Inflation with a singular potential

20

�p�g��1 L = � ap

2'p@µ'@µ'� C

's(1 +O('))

V =

8<

:C⇣

p�22pap�⌘ 2s

p�2+ · · · (p 6= 2),

Ces�/pap + · · · (p = 2).

ns =1� p+ s� 2

(p� 2)Nr =

8s

(p� 2)N

Potentials for monomial chaotic and power-law inflation

[Rinaldi, L. Vanzo, S. Zerbini, and G. Venturi, 1505.03386][TT, 1602.07867]

Summary of general pole inflation

p=1 1<p<2 p=2 2<p

2nd order hilltop

generalization of natural inflation

hilltop

alpha-attractor xi-attractor

Starobinsky model Higgs inflation

inverse-hilltop

run-away run-awaypower-law inflation

(exponential potential)monomial

chaotic

sing

ular

pot

entia

lno

n-si

ngul

ar p

oten

tial

For more details, see [TT, arXiv:1602.07867].

21

22

Correspondence to universality classes of inflation

General Pole Inflation As a realization of Universality Classes

Figures from [Garcia-Bellido, Roest, 1402.2059]

w/ log. corr.

w/ sing. pot.w/ sing. pot.

23

p > 2p < 2 p = 1p = 2 p = 2

p = 2p > 2

ここまでのまとめ

• インフレーション模型は Universality class に分類できる。

• インフラトン作用の極と次数による分類は、　　　　　　　　　それを実現する具体例となっている。

• さて、どのクラスが現実的でしょう？

24

Initial condition problems:Small-field or Large-field?

Figures from a review [Brandenberger, 1601.01918]

26

How likely the slow-roll is?Small field Large field

Inhomogeneous initial conditions

スカラー場の揺らぎがスローロールの領域を越えない限り非一様性が大きくてもインフレーションが起こる事を示した。

[East, Kleban, Linde, and Senatore, 1511.05143]

hr� ·r�i = 103⇤

結論small field: チューニングが必要

large field: robust

３＋１次元数値計算で

27

28

Universality classes of inflation[Garcia-Bellido et al., 1402.2059][Mukhanov, 1303.3925] [Roest, 1309.1285] [Binetruy et al., 1407.0820]

観測的ステータス

◎◯△×

△△

Any underlying mechanism for the universality?

初期条件

◎◎◎◎

×

Shift symmetry and its origin

Planck-suppressed terms are NOT suppressed enough!V ⇠ m2�2 + �3�

3 + �4�4 +

X

n>4

�4+n�4

✓�

MP

◆n

In particular,

V�2

M2P

⌘ = O(1)

Also a naturalness question:Why ? m ⌧ MP (or ⇤)

See e.g. a good review [Westphal, 1409.5350].

30

Shift symmetry� ! �+ c

with an explicit, soft breaking V (�) ⌧ 1

Perturbative quantum gravity corrections:

Technically natural.

[Kaloper, Lawrence, Sorbo, 1101.0026]

[Smolin, PLB 93, 95 (1980)][Linde, PLB 202 (1988) 194]

[’t Hooft, NATO Sci.Ser.B 59 (1980) 135]

31

�V ⇠ V

✓aV 00

M2P

+ bV

M4P

◆

Shift symmetry in SUGRASUSY breaking effects m

soft

⇠ O(H)

Why ( ) ?m ⌧ H ⌘ ⌧ 1

V = eK⇣K jiDiWDjW � 3|W |2

⌘+

1

2fABD

ADB

SUGRA scalar potential

DiW = Wi +KiWwhere .

32

shift symmetry

� ! �+ c K(�, �) = K(i(�� �))

[Kawasaki, Yamaguchi, Yanagida, hep-ph/0004243]

Origin of the shift symmetry?

33

U(1) RAxion Dilaton

pNG boson of U(1) orWilson line of extra dimensions pNG boson of scale invariance

shift symmetry … non-linearly realized symmetry … any linear realization?

Inverse-hilltop class と chaotic class は、このどちらかに帰着するべき？

[Freese, Frieman, Olinto, PRL 65 (1990) 3233]

U(1)-sym. inflation in SUGRA“Helical phase inflation”

34

[Li, Li, Nanopoulos, 1409.3267][Li, Li, Nanopoulos, 1412.5093][Li, Li, Nanopoulos, 1502.05005][Li, Li, Nanopoulos, 1507.04687]

[Ketov, TT, 1509.00953]

with the stabilizer field

without the stabilizer field

K =���� �2

0

�� ⇣

4

���� �2

0

�4.

W = m (c+ �)

cf.) with a monodromy

� =rp2ei✓/

p2�0

Extranatural inflationInflaton as an extra-dim. component of a gauge field

ei�/f = eiHA5dx

5

Discrete (gauged) shift symmetry: � ! �+ 2⇡nf

NOTE: The Wilson line is a non-local effect. Any local effects (including Quantum Gravity) cannot break the gauge symmetry explicitly.

[Arkani-Hamed et al., hep-th/0301218]

35

[Kaplan et al., hep-ph/0302014]

V ⇠X

n

cne�nS

cos

n�

f

Weak Gravity Conjecture“Gravity is the weakest force.”

For U(1) gauge theory with gravity, there must exist a particle satisfying

m < qMP

where m and q are the mass and U(1) charge of the particle.

Statement of the conjecture:

Application to a magnetic monopole:

The cut-off scale of the effective theory must satisfy

⇤ . gMP

where g is the gauge coupling (elementary charge).

36

[Arkani-Hamed et al., hep-th/0601001]

WGC: reasonIn limit, infinitely many charged black holes can exist.

The production probability will diverge even if each probability is exponentially suppressed.

(same as the argument against black hole remnants)However, there are objections to the arguments.See a recent review [Chen, Ong, Yeom, 1412.8366]

For the black holes to be able to decay, we need light enoughparticles.

g ! 0

MBH � QBHMP

(Extremal black holes saturate the bound.)

[Susskind, hep-th/9501106]

37

It also violates the covariant entropy bound conjecture [Bousso, hep-th/9905177].

WGC: generalization

T . gMP

For p-form gauge field, p-1-dim object satisfies

where T is the tension, and g is the charge density.

For 0-form (axion), S . MP

fwhere S is the instanton action, and f is the decay constant.

38

[Arkani-Hamed et al., hep-th/0601001]

(Extra)Natural inflation のまとめ

• discrete gauged shift symmetry に基づく魅力的なアイディア

• WGC により decay constant は厳しく制限される。（ループホールに注意。）

• WGC を回避できたとしても、既に Planck 1 σ 領域の外…。

See e.g. [Saraswat, 1608.06951] and references therein.

U(1)

39

Scale-invariant models of inflation

40

Starobinsky model

S =

Zd4x

p�g

✓R

2+

R

2

12M2

◆⇠

Zd4x

p�g

R

2

12M2

Higgs inflation

S =

Zd4x

p�g

✓R

2+ ⇠�

2R� 1

2(@µ�)

2 � ��

4

◆⇠

Zd4x

p�g

�⇣�

2R� ��

4�

gµ⌫ ! gµ⌫e2�

p�g !

p�ge4�

R ! Re�2�

� ! �e��

From a low-energy EFT point of view, see also [Csáki et al., 1406.5192].

[Starobinsky, PLB 91 (1980) 99]

[Bezrukov, Shaposhnikov, 0710.3755]

Unitarity violation or not

41

Provided that UV physics respects the asymptotic scale (shift) symmetry,the form of the Lagrangian is preserved by quantum corrections.

Taking into account the field dependence of the cut-off,the tree-level unitarity is preserved throughout the history of the universe.

[Bezrukov, Magnin, Shaposhnikov, Sibiryakov, 1008.5157]

Scale invariance in Swampland?Scale invariance はグローバル対称性なので、

量子重力効果で破れると期待される。

Any non-trivial constraints on EFT, like WGC?

Swampland

Landscape

[Vafa, hep-th/0509212] [Ooguri, Vafa, hep-th/0605264]

42

[Hawking, PRD 14 (1976) 2460]

Scale-invariant inflation のまとめ

• 観測と非常によく合っている。

• Global 対称性なので量子重力補正をコントロールできない。

• Swampland からの非自明な制限はあるか？（特に、摂動的ユニタリティーは量子重力的要求と両立するか？）

43

R

Summary & Conclusion

現実的なインフレーション模型は何か

観測的ステータス

◎◯△×

△△

初期条件

◎◎◎◎

×

シフト対称性の起源が不明瞭？

WGC を回避できれば現実的

データと合うが、 どう対称性を制御するか。 更なる研究が必要！45

…人間原理でチューニングしてフラットにする？ →何故まだ重力波が見つからない程フラットにできる？

Part II: Recent progress toward UV (SUGRA) embedding of inflation models

セミナーに呼んでください！