-
Notifications
You must be signed in to change notification settings - Fork 0
/
CRE70F3.tmp
5989 lines (5490 loc) · 351 KB
/
CRE70F3.tmp
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
\documentclass[a4paper,oneside,11pt,bibliography=totoc]{scrartcl}
%Use bibliography=totoc or bibliography=totocnumbered to show the List of References in the table of contents, numbered if using the latter
\usepackage[utf8]{inputenc}
\usepackage[T1]{fontenc}
%\usepackage{exscale}
%\newcommand\hmmax{0}
%\newcommand\bmmax{1}
%\newcommand{\Plus}{\mathord{\tikz\draw[line width=0.3ex, x=1ex, y=1ex] (0.75,0) -- (0.75,1.5)(0,0.75) -- (1.5,0.75);}}
\def\RPlus{\ensuremath{\mathbin{\rule[.13em]{.66em}{.22em}\hspace{-.44em}\rule[-.08em]{.22em}{.66em}\,}}} %Fat plus symbol
\usepackage[english]{babel}
\usepackage{graphicx} %extended \includegraphics options
%\usepackage{CJKutf8} %Kanji signs
\usepackage{dsfont}%even more mathematical symbols
%\usepackage{bm} %use \bm to get bold math symbols
\usepackage{amssymb} %even more mathematical symbols
\usepackage{amsthm} %theorem style usually needed for AMS (american mathematical society)
\usepackage{amsmath} %important stuff like equation structures and so on
\usepackage{amsfonts} %new mathematical symbols like fractals
%\usepackage{amscd}
%\usepackage{amstext}
%\usepackage{mathabx}
\usepackage{booktabs} %enhanced table stuff
%\usepackage{amsbsy} %bold mathematical symbols, also loaded in amsmath
\usepackage{fancybox} %for theorem boxes
%\usepackage[hypcap=false]{caption} %new captions for floating options
%\captionsetup{font={small,sf},labelfont=bf,textfont=rm}
%\usepackage{float} %provides H for images-here
%\usepackage{subfigure} %citing subfigures in a figure
\usepackage{lmodern} %latin modern font
\usepackage{fixcmex} %Fixing the issue with scaling brackets etc. in lmodern
%\renewcommand\familydefault{lmr}% uses lm only for text
%\usepackage[numbib,nottoc]{tocbibind} %Options for including the bibliography to content list etc.; nottoc prevents listing of Contents in the list of contents, numbib also numbers the section of the biblio
%\usepackage[pdfstartview={FitH},linkbordercolor={0 1 0}]{hyperref}
\usepackage[pdfstartview={FitH},linkbordercolor={0 1 0},colorlinks]{hyperref}
\usepackage{esint} %für \oiint und andere Integralformen
%\usepackage{fancyhdr} %Different handling for footers and headers
\usepackage[headsepline=.5pt,footsepline=.5pt,automark]{scrlayer-scrpage}
\usepackage{pdfpages} %For \includepdf etc
%\usepackage{trfsigns} % Fourier and Laplacesymbols for engineering und für \e für \exp
\newcommand{\e}{\ensuremath{\mathrm{e\;\!}}}
%\usepackage{wasysym} %Smiley
%\usepackage{siunitx} %SI Einheiten
%\usepackage{eqnarray} %Gleichungen in Tabellenform
\usepackage{mathtools} %für mathclap
\usepackage{relsize} %Integralgröße
\usepackage{scrhack} %Warnung bei book bezüglich toc
\usepackage{mleftright} %Distanz zu \left \right weg
\usepackage{tikz} %Diagramme
\usepackage{stmaryrd} % fuer llbracket usw.
\usepackage[many]{tcolorbox} %Coloured boxes around theorems etc
\tcbuselibrary{theorems} % siehe oben
%\usepackage{float} %For H descriptor in figure, but also for allowing figure in minipages
%\usepackage{xfrac} %for \sfrac quotient of sets
\usetikzlibrary{cd}
\makeatletter
\DeclareFontEncoding{LS1}{}{}
\DeclareFontSubstitution{LS1}{stix}{m}{n}
\DeclareMathAlphabet{\mathKel}{LS1}{stixscr}{m}{n}
\DeclareMathAlphabet{\mathcal}{LS1}{stixscr}{m}{n}
%Big tilde
\def\oversortoftilde#1{\mathop{\vbox{\m@th\ialign{##\crcr\noalign{\kern3\p@}%
\sortoftildefill\crcr\noalign{\kern3\p@\nointerlineskip}%
$\hfil\displaystyle{#1}\hfil$\crcr}}}\limits}
\def\sortoftildefill{$\m@th \setbox\z@\hbox{$\braceld$}%
\braceld\leaders\vrule \@height\ht\z@ \@depth\z@\hfill\braceru$}
\makeatother
\DeclareMathOperator{\sAut}{\mathKel{A\mkern-5.5mu u\mkern-4mu t\mkern-1.5mu}}
\DeclareMathOperator{\saut}{\mathKel{a\mkern-4.5mu u\mkern-4mu t\mkern-1.5mu}}
\DeclareMathOperator{\sEnd}{\mathKel{E\mkern-4mu n\mkern-4.5mu d\mkern-1mu}}
%\tikzset{every node/.style={align=center}}
%\usepackage{pst-node} %Diagramme
%\usepackage{auto-pst-pdf}
%\usepackage{stackengine} %Kommawedge
%\usepackage{mathabx}
%\usepackage{here}
%\usepackage[style=authortitle-icomp]{biblatex}
%\usepackage[babel,german=guillemets]{csquotes}
\setcounter{tocdepth}{4}
\setcounter{tocdepth}{5}
\setcounter{secnumdepth}{4}
\setcounter{secnumdepth}{5}
\setlength{\jot}{12pt} % Zeilenabstand in der Align-Umgebung
\allowdisplaybreaks % mit \displaybreak einen Seitenumbruch in align-Umgebung erzeugen
%\pagestyle{headings}
\pagestyle{scrheadings}
%\pagestyle{fancy} %eigener Seitenstil
%\fancyhf{}
%\lhead{\leftmark} %\nouppercase, um auch Kleinbuchstaben zu bekommen
%\renewcommand{\sectionmark}[1]{\markright{#1}{}}
% \markboth{\thesection{} #1}}
%\fancyhead[L]{Titel} %Kopfzeile links
\clearpairofpagestyles
\ihead{\headmark} %Kopfzeile links, automark has to be loaded in the package scrlayer-scrpage
\ohead{Simon-Raphael Fischer} %Kopfzeile rechts
%\renewcommand{\headrulewidth}{0.4pt} %obere Trennlinie
%\fancyfoot[C]{\thepage} %Seitennummer
\cfoot{\pagemark}
%\renewcommand{\footrulewidth}{0.4pt} %untere Trennlinie
\renewcommand{\theequation}{\arabic{equation}}
%\renewcommand*{\chapterpagestyle}{scrheadings}
%Folgender Befehl dafür da, dass sections auf ungeraden Seiten anfangen
%\let\Section\section
%\renewcommand\section[2][]{%
%\cleardoublepage
%\def\myTEMP{#1}\ifx\myTEMP\empty\Section{#2}\else\Section[#1]{#2}\fi}
%\usepackage[automark]{scrlayer-scrpage}
%\ohead{\pagemark}
%\ihead{\leftmark}
\renewcommand{\listoffigures}{\begingroup
\tocsection
\tocfile{\listfigurename}{lof}
\endgroup}
\renewcommand{\listoftables}{\begingroup
\tocsection
\tocfile{\listtablename}{lot}
\endgroup}
\oddsidemargin -0.0 cm
\evensidemargin -0.5 cm
\topmargin -1.5 cm
\headheight 1.2 cm
\headsep 1.3 cm
\topskip 0.5 cm
\textheight 23.0 cm
\textwidth 16.0 cm
\parindent 0.0 cm
\renewcommand{\baselinestretch}{1.2}
%\renewcommand{\arraystretch}{1.6}
\setcounter{totalnumber}{2}
\def\tp{^{^{^{\leftrightarrow}}}\!\!\!\!\!\!\!}
%%%%%%%%%%%%%%%%%%%%%%%%%%%% Definitionen %%%%%%%%%%%%%%%%%%%%%%%%%%%%
%\renewcommand{\chapterheadstartvskip}{\vspace{0cm}} %für bookklasse
\def\be{\begin{equation}}
\def\ee{\end{equation}}
\def\bs{\begin{subequations}}
\def\es{\end{subequations}}
\def\ba#1\ea{\begin{align}#1\end{align}}
\def\bes{\begin{equation*}}
\def\ees{\end{equation*}}
\def\bas#1\eas{\begin{align*}#1\end{align*}}
%um die Theoreme etc. in das Inhaltsverzeichnis zu packen
%\let\amsthmhead\thmhead
%\let\amsswappedhead\swappedhead
%\makeatletter
%\renewcommand*\thmhead[3]{\amsthmhead{#1}{#2}{#3}%
%%\@ifnotempty{#2}{\addcontentsline{toc}{subsection}{#2 ~ #3}}{}}
%\renewcommand*\swappedhead[3]{\amsswappedhead{#1}{#2}{#3}%
%%\@ifnotempty{#2}{\addcontentsline{toc}{subsection}{#2 ~ #3}}{}}
%\makeatother
\renewcommand{\qedsymbol}{$\blacksquare$}
\theoremstyle{plain}
\newtheorem{theorem}{Theorem}[section]
%\newtcbtheorem[number within=section]{theorem}{Theorem}
%{colback=green!5,colframe=green!35!black,fonttitle=\bfseries}{th}
\newtheorem{corollary}[theorem]{Corollary}
\newtheorem{lemma}[theorem]{Lemma}
\newtcbtheorem
[use counter*=theorem,number within=section]% init options
{lemmata}% name
{Lemma}% title
{%
fontupper=\itshape,
breakable,
enhanced,
colback=gray!25,
colframe=gray!0!black,
fonttitle=\bfseries,
}% options
{lem}% prefix
\newtheorem{proposition}[theorem]{Proposition}
\newtcbtheorem
[use counter*=theorem,number within=section]% init options
{propositions}% name
{Proposition}% title
{%
fontupper=\itshape,
breakable,
enhanced,
colback=gray!25,
colframe=gray!0!black,
fonttitle=\bfseries,
}% options
{prop}% prefix
\newtcbtheorem
[use counter*=theorem,number within=section]% init options
{theorems}% name
{Theorem}% title
{%
fontupper=\itshape,
breakable,
enhanced,
colback=gray!25,
colframe=gray!0!black,
fonttitle=\bfseries,
}% options
{thm}% prefix
\newtcbtheorem
[use counter*=theorem,number within=section]% init options
{corollaries}% name
{Corollary}% title
{%
fontupper=\itshape,
breakable,
enhanced,
colback=gray!25,
colframe=gray!0!black,
fonttitle=\bfseries,
}% options
{cor}% prefix
\newtheorem{conjecture}[theorem]{Conjecture}
\theoremstyle{remark}
\newtcbtheorem
[use counter*=theorem,number within=section]% init options
{remarks}% name
{Remark}% title
{%
breakable,
enhanced,
colback=gray!5,
colframe=gray!50!black,
fonttitle=\bfseries,
}% options
{rem}% prefix
\newtheorem*{note}{Note}
\newtheorem{remark}[theorem]{Remarks}
\newtheorem{motivation}[theorem]{Motivation}
\theoremstyle{definition}
\newtheorem{definition}[theorem]{Definition}
\newtcbtheorem
[use counter*=theorem,number within=section]% init options
{definitions}% name
{Definition}% title
{%
breakable,
enhanced,
colback=gray!25,
colframe=gray!0!black,
fonttitle=\bfseries,
}% options
{def}% prefix
\newtcbtheorem
[use counter*=theorem,number within=section]% init options
{examples}% name
{Example}% title
{%colback=black!5,colframe=red!35!black
breakable,
enhanced,
colback=gray!5,
colframe=gray!50!black,
fonttitle=\bfseries,
}% options
{ex}% prefix
\newtcbtheorem
[use counter*=theorem,number within=section]% init options
{situations}% name
{Situation}% title
{%colback=black!5,colframe=red!35!black
breakable,
enhanced,
colback=gray!5,
colframe=gray!50!black,
fonttitle=\bfseries,
}% options
{sit}% prefix
\newtcbtheorem
[use counter*=theorem,number within=section]% init options
{fieldredefinitions}% name
{Theorem}% title
{%colback=black!5,colframe=red!35!black
breakable,
enhanced,
colback=gray!25,
colframe=gray!0!black,
fonttitle=\bfseries,
}% options
{fieldredef}% prefix
\newtheorem{example}[theorem]{Example}
% hier Namen etc. einsetzen
%\newcommand{\fullname}{Simon-Raphael Fischer}
%\newcommand{\email}{sfischer@ncts.tw}
%\newcommand{\titel}{Examples and No-Gos of curves Yang-Mill-Higgs gauge theories}
%\newcommand{\titel}{Titel der Arbeit}
%\newcommand{\jahr}{2020}
%\newcommand{\matnr}{11424951}
%\newcommand{\gutachterA}{Prof. Dr. Mark John David Hamilton}
%\newcommand{\betreuer}{Professor Dr. Anna Dall'Acqua}
% hier richtige Fakultät auswählen
%\newcommand{\fakultaet}{Fakultät noch ergänzen}
%\newcommand{\fakultaet}{Mathematik und\\Wirtschaftswissenschaften}
%\newcommand{\fakultaet}{Naturwissenschaften}
%\newcommand{\fakultaet}{Medizin}
% nun noch unten das Institut einsetzen
%\newcommand{\institut}{Institut noch ergänzen}
% Für \widecheck: (umgekehrter Zirkumflex)
\DeclareFontFamily{U}{mathx}{\hyphenchar\font45}
\DeclareFontShape{U}{mathx}{m}{n}{
<5> <6> <7> <8> <9> <10>
<10.95> <12> <14.4> <17.28> <20.74> <24.88>
mathx10
}{}
\DeclareSymbolFont{mathx}{U}{mathx}{m}{n}
\DeclareFontSubstitution{U}{mathx}{m}{n}
\DeclareMathAccent{\widecheck}{0}{mathx}{"71}
\DeclareMathAccent{\wideparen}{0}{mathx}{"75}
%\newtheorem{cor}{Corollary}
%\newtheorem{ad}{Theorem}
%\newtheorem{theorem}{Theorem}
%\newtheorem{theorem}{Theorem}
%\newtheorem{theorem}{Theorem}
%\includeonly{PhysicalBasics/f(R)gravity} %nur das kompilieren
%\let\endtitlepage\relax %No page break after titlepage
\begin{document}
\pagenumbering{Roman}
\renewcommand{\thefootnote}{\fnsymbol{footnote}}
\begin{titlepage}
%\thispagestyle{empty}
\author{Simon-Raphael Fischer\footnote{Email: \href{mailto:sfischer@ncts.tw}{sfischer@ncts.tw}; ORCiD: \href{https://orcid.org/0000-0002-5859-2825}{0000-0002-5859-2825}} }
\title{Integrating curved Yang-Mills gauge theories}
\subtitle{Gauge theories related to principal bundles equipped with Lie group bundle actions}
\date{\today}
\maketitle
\thispagestyle{empty}
\begin{center}
National Center for Theoretical Sciences, Mathematics Division, National Taiwan University\\
No. 1, Sec. 4, Roosevelt Rd., Taipei City 106, Taiwan Room 503, Cosmology Building, Taiwan
\ \\
\ \\
\ \\
\textbf{Abstract}\footnote[2]{Abbreviations used in this paper: \textbf{LGB} for Lie group bundle, \textbf{LAB} for Lie algebra bundle.}
%\footnote[2]{Abbreviations used in this paper: \textbf{(C)YMH GT} for (curved) Yang-Mills-Higgs gauge theory.}
\begin{abstract}
\small{
In this paper we construct a gauge theory based on principal bundles $\mathcal{P}$ equipped with a Lie group bundle action instead of a Lie group action. Due to the fact that pushforwards of right-translation act now only on the vertical structure of $\mathcal{P}$ we fix a connection on $\mathcal{G}$, which modifies the pushforward via right translation by subtracting the fundamental vector field generated by the Darboux derivative of a section of $\mathcal{G}$.
}
\end{abstract}
\end{center}
\textit{2020 MSC:} Primary 53D17; Secondary 81T13, 17B99.
\textit{Keywords:} \texttt{Mathematical Gauge Theory}, Differential Geometry, High Energy Physics - Theory, Mathematical Physics
\end{titlepage}
%%%%%%%%%%%%%%%%%%%%%%%%%%%% Inhaltsverzeichnis %%%%%%%%%%%%%%%%%%%%%%%%%%%%
\tableofcontents
%\thispagestyle{empty}\hspace{1em}\newpage
% \thispagestyle{empty}\hspace{1em}\newpage
\renewcommand{\thefootnote}{\arabic{footnote}}
%
\setlength{\parindent}{12 pt}
%%%%%%%%%%%%%%%%%%%%%%%%%%%% Hier beginnt der Hauptteil %%%%%%%%%%%%%%%%%%%%%%%%%%%%
%\cleardoublepage
% für die leere(n) Seite(n)
\pagenumbering{arabic}
\section{Introduction}
\subsection{Basic notations and remarks}\label{BasicNotations}
\begin{itemize}
\item The appendix serves for providing extra information or background knowledge which did not fit in the flow of this paper's text. We will sometimes refer to the appendix, but the experienced reader may be able to ignore the appendix.
\item In the main text we usually repeat and reintroduce needed objects for the statements (like "\textbf{Let $M$ be a manifold [...]}" in every statement) (almost) allowing to just read the statements without having read the text introducing it, while the appendix is written as a continuous text which has to be read as a whole in order to understand the essential statements.
\item With $f^*F$ we denote the pullback/pull-back of the fibre bundles $F \to M$ under a smooth map $f: N \to M$. Similarly we denote the pullbacks of sections of a fibre bundle.
%We will also have sections $F$ as an element of $\Gamma\left( \left(\bigotimes_{m=1}^{l} E_m^*\right) \otimes E_{l+1} \right)$, where $E_1, \dots E_{l+1}$ ($l \in \mathbb{N}$) are real vector bundles of finite rank over $N$. Those pull-back as section, denoted by $\Phi^*F$, we will view as an element of $\Gamma\left( \mleft(\bigotimes_{m=1}^{l} \mleft(\Phi^*E_m\mright)^*\mright) \otimes \Phi^*E_{l+1} \right)$, and it is essentially given by
%\bas
%(\Phi^*F)(\Phi^*\nu_1, \dotsc , \Phi^*\nu_l)
%&=
%\Phi^*\mleft( F\mleft( \nu_1, \dotsc, \nu_l \mright) \mright)
%\eas
%for all $\nu_1 \in \Gamma(E_1), \dotsc, \nu_l \in \Gamma(E_l)$, using that pullbacks of sections generate the sections of a pullback bundle.
\item For $V \to M$ a vector bundle over $M$ do not confuse the pull-back of sections with the pull-back of forms $\omega \in \Omega^l(M; V)$ ($l \in \mathbb{N}_0$), here denoted by $f^!\omega$, which is an element of $\Gamma\left( \mleft(\bigwedge_{m=1}^{l} \mathrm{T}^*M \mright) \otimes f^*V \right) \cong \Omega^l(M; f^*V)$, and not of $\Gamma\left( \mleft(\bigotimes_{m=1}^{l} \mleft(f^*\mathrm{T}N\mright)^*\mright) \otimes f^*V \right)$ like $f^*\omega$.
%$f^!\omega$ is defined by
%\ba
%\mleft.\mleft(f^!\omega\mright)(Y_1, \dots, Y_l)\mright|_p
%&\coloneqq
%\omega_{f(p)}\mleft(\mathrm{D}_pf\mleft(\mleft.Y_1\mright|_p\mright), \dots, \mathrm{D}_pf\mleft(\mleft.Y_l\mright|_p\mright)\mright)
%\ea
%for all $p \in M$ and $Y_1, \dots, Y_l \in \mathfrak{X}(M)$.
\item Let $F \stackrel{\pi_F}{\to} M$ and $G \stackrel{\pi_G}{\to} N$ be two fibre bundles over smooth manifolds $M$ and $N$, and let $\phi: N \to M$ be a smooth map. Furthermore, let us assume we have a morphism $\Phi: G \to F$ of fibre bundles over $\phi$, that is, $\Phi$ is a smooth map such that the following diagram commutes
\begin{center}
\begin{tikzcd}
G \arrow{d}{\pi_G} \arrow{r}{\Phi} & F \arrow{d}{\pi_F}\\
N \arrow{r}{\phi} & M
\end{tikzcd}
\end{center}
especially, $\pi_F \circ \Phi = \phi \circ \pi_G$.
We make often use of that such morphisms have a 1:1 correspondence to \textbf{base-preserving} fibre bundle morphisms $\widetilde{\Phi}: G \to \phi^*F$, \textit{i.e.}\ $\widetilde{\Phi}$ is a smooth map with $\phi^*\pi_F \circ \widetilde{\Phi} = \pi_G$. For $p \in N$ the morphism $\widetilde{\Phi}$ has the form
\bas
\widetilde{\Phi}_p
&=
(p, \Phi_p),
\eas
that is,
\bas
\widetilde{\Phi}_p(g)
&=
\bigl(p, \Phi_p(g) \bigr)
\eas
for all $g \in G_p$, which is well-defined since $\Phi_p(g) \in F_{\phi(p)}$. The map $\widetilde{\Phi} \mapsto \Phi \coloneqq \mathrm{pr}_2 \circ \widetilde{\Phi}$ is then a bijective map between base-preserving morphisms $G \to \phi^*F$ and morphisms $G \to F$ over $\phi$, where $\mathrm{pr}_2$ is the projection onto the second component.
In total, $\widetilde{\Phi}$ is a base-preserving morphism if and only if $\Phi$ is a morphism over $\phi$; in fact, one defines pullback bundles in such a way that this equivalence holds. Observe that $\widetilde{\Phi}$ is an isomorphism (diffeomorphism) if and only if $\Phi$ is a fibre-wise isomorphism (diffeomorphism).
One can extend all of this similarly for more specific types of morphisms like vector bundle-morphisms.
Most of the time we will not mention this 1:1 correspondence explicitly, it should be clear by context. Hence, we will also denote $\widetilde{\Phi}$ by $\Phi$. In fact, we usually calculate with $\widetilde{\Phi}$, while $\Phi$ and its diagram may only arise to give an illustration about the geometry.
\item When we differentiate maps $\gamma$ depending on just one parameter $t \in \mathbb{R}$, then we may write
\bas
\frac{\mathrm{d}}{\mathrm{d}t} \gamma(t)
&\coloneqq
\mleft.\frac{\mathrm{d}}{\mathrm{d}t} [t \mapsto \gamma]\mright|_t.
\eas
\end{itemize}
\subsection{Assumed background knowledge}
It is highly recommended to have basic knowledge about differential geometry and gauge theory as presented in \cite[especially Chapter 1 to 5]{Hamilton}, and we will follow the style and labeling as in \cite{Hamilton} when we generalize certain notions; however, sometimes we will still give explicit references to help with more technical details. It can be useful to have knowledge about Lie algebra and Lie group bundles, and even Lie algebroids and Lie groupoids, but we will introduce their basic notions such that it is not necessarily needed to have knowledge about these upfront.
We also often give references about Lie group bundles (LGBs), but the given references are often about Lie groupoids. If the reader has no knowledge about Lie groupoids, then it is important to know that LGBs are a special example of Lie groupoids; Lie groupoids carry "two projections", called \textbf{source} and \textbf{target}. An LGB is a special example of a Lie groupoid whose source equals the target.\footnote{But not every Lie groupoid with equal source and target is an LGB, they're in general bundles of Lie groups which is not completely the same; this nuance will not be important here.} If you look into such a reference, then the source and target are often denoted by $\alpha$ and $\beta$, or by $s$ and $t$; simply put both to be the same and identify these with our bundle projection which we often denote by $\pi$. In that way it should be possible to read the references without the need to know Lie groupoids. However, we try to re-prove the needed statements such that these types of references could be avoided by the reader.
See also the previous subsection about notions we assume to be known.
%\twocolumn
\section{Basic definitions}\label{BasicDefinitions}
%
%In the following, we denote with $V^*$ the dual of a vector bundle $V \to N$ over a smooth manifold $N$, and $\Phi^*V$ denotes the pull-back of $V$ by $\Phi: M \to N$, a smooth map from a smooth manifold $M$ to $N$. We have a similar notation for the pull-back of sections, especially we will have sections $F$ as an element of $\Gamma\left( \left(\bigotimes_{m=1}^{l} E_m^*\right) \otimes E_{l+1} \right)$, where $E_1, \dots, E_{l+1} \to N$ ($l \in \mathbb{N}$) are real vector bundles of finite rank over a smooth manifold $N$, and $\Gamma(\cdot)$ denotes the space of smooth sections. Then we view the pull-back $\Phi^*F$ as an element of $\Gamma\left( \mleft(\bigotimes_{m=1}^{l} \mleft(\Phi^*E_m\mright)^*\mright) \otimes \Phi^*E_{l+1} \right)$, and it is essentially given by
%\bas
%(\Phi^*F)(\Phi^*\nu_1, \dotsc , \Phi^*\nu_l)
%&=
%\Phi^*\mleft( F\mleft( \nu_1, \dotsc, \nu_l \mright) \mright)
%\eas
%for all $\nu_1 \in \Gamma(E_1), \dotsc, \nu_l \in \Gamma(E_l)$. In general we also make use of that sections of $\Phi^*E$ can be viewed as sections of $E$ along $\Phi$, where $E \stackrel{\pi}{\to} N$ is any vector bundle over $N$. Let $\mu \in \Gamma(\Phi^*E)$, then it has the form $\mu_p = (p, u_p)$ for all $p \in M$, where $u_p \in E_{\Phi(p)}$, the fibre of $E$ at $\Phi(p)$; and a section $\nu$ of $E$ along $\Phi$ is a smooth map $M \to E$ such that $\pi \circ \nu = \Phi$. Then on one hand $\mathrm{pr}_2 \circ \mu$ is a section along $\Phi$, where $\mathrm{pr}_2$ is the projection onto the second component, and on the other hand $M \ni p \mapsto (p, \nu_p)$ defines an element of $\Gamma(\Phi^*E)$. With that one can show that there is a 1:1 correspondence of $\Gamma(\Phi^*E)$ with sections along $\Phi$. Similarly, vector bundle morphisms $L: G \to E$ over $\Phi$ have 1:1 correspondences to base-preserving vector bundle morpishms $G \to \Phi^*E$, where $G \to M$ is a vector bundle over $M$. We do not necessarily mention it when we make use of such trivial identifications, it should be clear by the context. For example $\mathrm{D}\Phi$ denotes the total differential of $\Phi$ (also called tangent map). It can be viewed as a vector bundle morphism $\mathrm{T}M \to \mathrm{T}N$ over $\Phi$, and we often view it as an element of $\Omega^1(M; \Phi^*\mathrm{T}N)$ by $\mathfrak{X}(M) \ni Y \mapsto \mathrm{D}\Phi(Y)$, where $\mathrm{D}\Phi(Y) \in \Gamma(\Phi^*\mathrm{T}N), M \ni p \mapsto \mathrm{D}_p\Phi(Y_p)$.
%
%Additionally, with $\Omega^k(N; E)$ ($k \in \mathbb{N}_0$) we denote $k$-forms on $N$ with values in a vector bundle $E \to N$, and we always use the Einstein's sum convention. If one has a connection $\nabla$ on a vector bundle $V \to N$, then one has the notion of the exterior covariant derivative on $\Omega^p(M;E)$, denoted by $\mathrm{d}^\nabla$. In the case of a trivial vector bundle $V=N \times W \to N$, where $W$ is some vector space, we will often use the \textbf{canonical flat connection} for $\nabla$, defined by $\nabla \nu = 0$, where $\nu$ is a constant section of $N \times W$, see \textit{e.g.}~\cite[Example 5.1.7; page 260f.]{hamilton}\ for a geometric interpretation as horizontal distribution. The canonical flat connection is clearly uniquely defined (if a trivialization is given) because constant sections generate all sections and due to the Leibniz rule and linearity of $\nabla$. Let $\mleft( e_a \mright)_a$ be a constant global frame of $N \times W$, thence,
%\bas
%\mathrm{d}^\nabla \omega
%&=
%\mathrm{d} \omega^a \otimes e_a
%\eas
%for all $\omega \in \Omega^p(M; W)$, where we write $\omega= \omega^a \otimes e_a$. Hence, we define
%\ba
%\mathrm{d}\omega
%&\coloneqq
%\mathrm{d}^\nabla \omega,
%\ea
%when $\nabla$ is the canonical flat connection. $\mathrm{d}$ is clearly a differential.
%
%As usual, there will be definitions of certain objects depending on other elements, and for keeping notations simple we will not always explicitly denote all dependencies. It will be clear by context on which it is based on, that is, when we define an object $A$ using the notion of Lie algebra actions $\gamma$ and we write "Let $A$ be [as defined before]", then it will be clear by context which Lie algebra action is going to be used, for example given in a previous sentence writing "Let $\gamma$ be a Lie algebra action".
%%, and recall the following wedge product\footnote{As also defined in \cite[\S 5, third part of Exercise 5.15.12; page 316]{hamilton}.} of forms with values in a vector bundle $E$ and values in its space of endomorphisms $\mathrm{End}(E)$,
%%\bas
%%\wedge: \Omega^k(N; \mathrm{End}(E)) \times \Omega^l(N; E)
%%&\mapsto
%%\Omega^{k+l}(N; E) \\
%%(T, \omega) &\mapsto T \wedge \omega
%%\eas
%%for all $k, l \in \mathbb{N}_0$, given by
%%\ba\label{DefVonWedgedemitEnd}
%%\mleft( T \wedge \omega \mright) \mleft( Y_1, \dotsc, Y_{k+l} \mright)
%%&\coloneqq
%%\frac{1}{k! l!} \sum_{\sigma \in S_{k+l}} \mathrm{sgn}(\sigma) ~
%%T \mleft( Y_{\sigma(1)}, \dotsc, Y_{\sigma(k)} \mright)
%%\mleft( \omega\mleft( Y_{\sigma(k+1)}, \dotsc, Y_{\sigma(k+l)} \mright) \mright),
%%\ea
%%where $S_{k+l}$ is the group of permutations $\{1, \dotsc, k+l\}$. This is then locally given by, with respect to a frame $\mleft( e_a \mright)_a$ of $E$,
%%\bas
%%T \wedge \omega &= T(e_a) \wedge w^a,
%%\eas
%%where $T$ acts as an endomorphism on $e_a$, \textit{i.e.}~$T(e_a) \in \Omega^k(N; E)$, and $\omega = \omega^a \otimes e_a$. Also recall that there is the canonical extension of $\nabla$ on $\mathrm{End}(E)$ by forcing the Leibniz rule. We still denote this connection by $\nabla$, too.
%
%We also need the following definitions.
%
%\begin{definitions}{Graded extension of products, \newline \cite[generalization of Definition 5.5.3; page 275]{hamilton}}{GradingOfProducts}
%Let $l \in \mathbb{N}$ and $E_1, \dots E_{l+1} \to N$ be vector bundles over a smooth manifold $N$, and $F \in \Gamma\left( \left(\bigotimes_{m=1}^{l} E_m^*\right) \otimes E_{l+1} \right)$. Then we define the \textbf{graded extension of $F$} as
%\bas
%\Omega^{k_1}(N; E_1) \times \dots \times \Omega^{k_l}(N; E_l)
%&\to \Omega^{k}(N; E_{l+1}), \\
%(A_1, \dots, A_l)
%&\mapsto
%F\mleft(A_1\stackrel{\wedge}{,} \dotsc \stackrel{\wedge}{,} A_l\mright),
%\eas
%where $k := k_1+\dots k_l$ and $k_i \in \mathbb{N}_0$ for all $i\in \{1, \dots, l\}$. $F\mleft(A_1\stackrel{\wedge}{,} \dotsc \stackrel{\wedge}{,} A_l\mright)$ is defined as an element of $\Omega^{k}(N; E_{l+1})$ by
%\bas
%&F\mleft(A_1\stackrel{\wedge}{,} \dotsc \stackrel{\wedge}{,} A_l\mright)\mleft(Y_1, \dots, Y_{k}\mright)
%\coloneqq \\
%&\frac{1}{k_1! \cdot \dots \cdot k_l!} \sum_{\sigma \in S_{k}} \mathrm{sgn}(\sigma) ~ F\left( A_1\left( Y_{\sigma(1)}, \dots, Y_{\sigma(k_1)} \right), \dots, A_l\left( Y_{\sigma(k-k_l+1)}, \dots, Y_{\sigma(k)} \right) \right)
%\eas
%for all $Y_1, \dots, Y_{k} \in \mathfrak{X}(N)$, where $S_{k}$ is the group of permutations of $\{1, \dots, k\}$ and $\mathrm{sgn}(\sigma)$ the signature of a given permutation $\sigma$.
%
%$\stackrel{\wedge}{,}$ may be written just as a comma when a zero-form is involved.
%
%Locally, with respect to given frames $\mleft( e^{(i)}_{a_i} \mright)_{a_i}$ of $E_i$, this definition has the form
%\ba\label{CoordExprOfGradedExtension}
%F\mleft(A_1\stackrel{\wedge}{,} \dotsc \stackrel{\wedge}{,} A_l\mright)
%&=
%F\mleft(e^{(1)}_{a_1}, \dotsc, e^{(l)}_{a_l}\mright) \otimes A_1^{a_1} \wedge \dotsc \wedge A_l^{a_l}
%\ea
%for all $A_i = A_i^{a_i} \otimes e^{(i)}_{a_i}$, where $A_i^{a_i}$ are $k_i$-forms on $N$.
%\end{definitions}
%
%\begin{remark}
%\leavevmode\newline
%Assume $F \in \Gamma\left( \mleft(\bigwedge_{m=1}^{l} \mathrm{T}^*N \mright) \otimes E \right) \cong \Omega^l(N; E)$ for some vector bundle $E$, \textit{i.e.}~$F$ is an $l$-form on $N$ with values in $E$. The pull-back $\Phi^*F$ by $\Phi$ can be then viewed as an element of $\Gamma\left( \bigwedge_{m=1}^{l} \mleft(\Phi^*\mathrm{T}N\mright)^* \otimes \Phi^*E \right)$.
%
%Do not confuse this pull-back with the pull-back of forms, here denoted by $\Phi^!F$, which is an element of $\Gamma\left( \mleft(\bigwedge_{m=1}^{l} \mathrm{T}^*M \mright) \otimes \Phi^*E \right) \cong \Omega^l(M; \Phi^*E)$ defined by
%\ba
%\mleft.\mleft(\Phi^!F\mright)(Y_1, \dots, Y_l)\mright|_p
%&\coloneqq
%F_{\Phi(p)}\mleft(\mathrm{D}_p\Phi\mleft(\mleft.Y_1\mright|_p\mright), \dots, \mathrm{D}_p\Phi\mleft(\mleft.Y_l\mright|_p\mright)\mright)
%\ea
%for all $p \in M$ and $Y_1, \dots, Y_l \in \mathfrak{X}(M)$. Then
%\ba\label{EqPullBackFormelFuerVerschiedeneDefinitionen}
%\Phi^!F
%&=
%\frac{1}{l!}~
%\mleft(\Phi^*F\mright) ( \underbrace{\mathrm{D}\Phi \stackrel{\wedge}{,} \dotsc \stackrel{\wedge}{,} \mathrm{D}\Phi}_{l \text{ times}} )
%\ea
%by using the anti-symmetry of $F$ and Def.~\ref{def:GradingOfProducts}, \textit{i.e.}
%\bas
%&\mleft.\frac{1}{l!}~
%\Big(\mleft(\Phi^*F\mright) ( \mathrm{D}\Phi \stackrel{\wedge}{,} \dotsc \stackrel{\wedge}{,} \mathrm{D}\Phi ) \Big) (Y_1, \dots, Y_l)\mright|_p \\
%&\hspace{1cm}
%=
%\frac{1}{l!}~
%\sum_{\sigma \in S_{l}} \mathrm{sgn}(\sigma) ~ \underbrace{(\Phi^*F)\mleft(\mathrm{D}\Phi\mleft(Y_{\sigma(1)}\mright), \dots, \mathrm{D}\Phi\mleft(Y_{\sigma(l)}\mright)\mright)}_{\mathclap{= \mathrm{sgn}(\sigma) ~ (\Phi^*F)\mleft(\mathrm{D}\Phi\mleft(Y_{1}\mright), \dots, \mathrm{D}\Phi\mleft(Y_{l}\mright)\mright)}}\Big|_p \\
%&\hspace{1cm}
%=
%\frac{1}{l!}~ \underbrace{\mleft( \sum_{\sigma \in S_{l}} 1 \mright)}_{= l!} ~
%F_{\Phi(p)}\mleft(\mathrm{D}_p\Phi\mleft(\mleft.Y_{1}\mright|_p\mright), \dots, \mathrm{D}_p\Phi\mleft(\mleft.Y_{l}\mright|_{p}\mright)\mright) \\
%&\hspace{1cm}
%= \mleft.\mleft(\Phi^!F\mright)(Y_1, \dots, Y_l)\mright|_p
%\eas
%for all $p \in M$ and $Y_1, \dots, Y_l \in \mathfrak{X}(M)$.
%\end{remark}
%
%In case of antisymmetric tensors we of course preserve that.
%
%\begin{propositions}{Graded extensions of antisymmetric tensors}{GradedExtensionPlusAntiSymm}
%Let $E_1, E_2 \to N$ be real vector bundles of finite rank over a smooth manifold $N$, $F \in \Omega^2(E_1; E_2)$. Then
%\ba
%F \mleft( A \stackrel{\wedge}{,} B \mright)
%&=
%-\mleft( -1 \mright)^{km}
%F \mleft( B \stackrel{\wedge}{,} A \mright)
%\ea
%for all $A \in \Omega^k(N; E_1)$ and $B \in \Omega^m(N; E_2)$ ($k,m \in \mathbb{N}_0$). Similarly extended to all $F \in \Omega^l(E_1; E_2)$.
%\end{propositions}
%
%\begin{remark}
%\leavevmode\newline
%This is a generalization of similar relations just using the Lie algebra bracket $\mleft[ \cdot, \cdot\mright]_{\mathfrak{g}}$ of a Lie algebra $\mathfrak{g}$, see \cite[\S 5, first statement of Exercise 5.15.14; page 316]{hamilton}.
%\end{remark}
%
%\begin{proof}
%\leavevmode\newline
%Trivial by using Eq.~\eqref{CoordExprOfGradedExtension}.
%\end{proof}
%
%We also need to know what a Lie algebroid is, a generalization of both, tangent bundles and Lie algebras; this concept will just be defined, refer to the references for thorough discussions of these definitions, especially \cite{mackenzieGeneralTheory} and \cite[\S VII; page 113ff.]{DaSilva}.
%
%\begin{definitions}{Lie algebroid, \newline \cite[\S 3.3, first part of Definition 3.3.1; page 100]{mackenzieGeneralTheory}}{test}
%%\leavevmode\newline
%Let $E \to N$ be a real vector bundle of finite rank. Then $E$ is a smooth Lie algebroid if there is a bundle map $\rho: E \to \mathrm{T}N$, called the \textbf{anchor}, and a Lie algebra structure on $\Gamma(E)$ with Lie bracket $\mleft[ \cdot, \cdot \mright]_E$ satisfying
%\ba
%\mleft[\mu, f \nu\mright]_E = f \mleft[\mu, \nu\mright]_E + \mathcal{L}_{\rho(\mu)}(f) ~ \nu
%\label{eq:E-Leibniz}
%\ea
%for all $f \in C^\infty(N)$ and $\mu, \nu \in \Gamma(E)$, where $\mathcal{L}_{\rho(\mu)}(f)$ is the action of the vector field $\rho(\mu)$ on the function $f$ by derivation. We will sometimes denote a Lie algebroid by $\mleft( E, \rho, \mleft[ \cdot, \cdot \mright]_E \mright)$.
%%We will sometimes denote a Lie algebroid by $\mleft( E, \rho, \mleft[ \cdot, \cdot \mright]_E \mright)$.
%\end{definitions}
%
%%Tangent bundles are a canonical example of Lie algebroids, their anchor is the identity with which we also equip them; another canonical example with zero anchor are the Lie algebra bundles:
%Tangent bundles and bundles of Lie algebras are canonical examples of Lie algebroids, their anchor is the identity and zero, respectively. The important example for us is a mixture of those examples:
%
%\begin{propositions}{Action Lie algebroids, \cite[\S 16.2, Example 5; page 114]{DaSilva}}{ActionLieoidsAreOids}
%Let $\mleft(\mathfrak{g}, \mleft[\cdot, \cdot \mright]_{\mathfrak{g}}\mright)$ be some Lie algebra equipped with a Lie algebra action $\gamma: \mathfrak{g} \to \mathfrak{X}(N)$ on a smooth manifold $N$. Then there is a unique Lie algebroid structure on $E = N \times \mathfrak{g}$ such that we have
%\ba
%\rho(\nu)
%&=
%\gamma(\nu),
%\\
%\mleft[\mu, \nu\mright]_E
%&=
%\mleft[\mu, \nu\mright]_{\mathfrak{g}}
%\ea
%for all constant sections $\mu, \nu \in \Gamma(E)$. We call this structure \textbf{action Lie algebroid}.
%\end{propositions}
\section{Lie group bundles (LGBs)}
\subsection{Definition}
%\pagebreak
\begin{definitions}{Lie group bundle, \cite[\S 1.1, Def.\ 1.1.19; p. 11]{mackenzieGeneralTheory}}{LieGroupBundle}
Let $G, \mathcal{G}, M$ be smooth manifolds. A fibre bundle
\begin{center}
\begin{tikzcd}
G \arrow{r} & \mathcal{G} \arrow{d}{\pi} \\
& M
\end{tikzcd}
\end{center}
is called a \textbf{Lie group bundle} if:
\begin{enumerate}
\item $G$ and each fibre $\mathcal{G}_x \coloneqq \pi^{-1}\mleft( \{x\} \mright)$, $x\in M$, are Lie groups;
\item there exists a bundle atlas $\mleft\{ \mleft( U_i, \phi_i \mright) \mright\}_{i \in I}$ such that the induced maps
\bas
\phi_{ix}
&\coloneqq
\mathrm{pr}_2 \circ \mleft. \phi_i\mright|_{\mathcal{G}_x}: \mathcal{G}_x \to G
\eas
are Lie group isomorphisms, where $I$ is an (index) set, $U_i$ are open sets covering $M$, $\phi_i: \mathcal{G}|_U \to U \times G$ subordinate trivializations, and $\mathrm{pr}_2$ the projection onto the second factor. This atlas will be called \textbf{Lie group bundle atlas} or \textbf{LGB atlas}.
\end{enumerate}
We also often say that \textbf{$\mathcal{G}$ is an LGB (over $M$)}, whose structural Lie group is either clear by context or not explicitly needed; and we may also denote LGBs by $G \to \mathcal{G} \stackrel{\pi}{\to} M$.
\end{definitions}
\begin{remarks}{Principal and Lie group bundles}{LiegroupbundlesNotPrincipalBundles}
Beware, a Lie group bundle is \textbf{not} the same as a principal bundle $P \to M$ with the same fibre type $G$. First of all, the fibres of $P$ are just diffeomorphic to a Lie group, a priori they carry no Lie group structure, while the fibres of $\mathcal{G}$ carry a Lie group structure.
\newline
Second, on $P$ we have a multiplication given as an action of $G$ on $P$
\bas
P \times G \to P,
\eas
preserving the fibres $P_x$ ($x\in M$) and simply transitive on them. Restricted on $P_x$ we have
\bas
P_x \times G \to P_x.
\eas
For $\mathcal{G}$ we have canonically a multiplication over $x$ given by
\bas
\mathcal{G}_x \times \mathcal{G}_x \to \mathcal{G}_x,
\eas
also clearly simply transitive. Observe, the second factor is not "constant", \textit{i.e.}\ we do not have $\mathcal{G}_x \times G \to \mathcal{G}_x$ in general. Hence, there is in general no well-defined product $\mathcal{G} \times \mathcal{G} \to \mathcal{G}$.
\newline
All of that is also resembled in the existence of sections. The existence of a section of $P$ has a 1:1 correspondence to trivializations of $P$, which is why $P$ in general only admits sections locally; see \textit{e.g.}\ \cite[\S 4.2, Thm.\ 4.2.19; page 219f.]{Hamilton}. $\mathcal{G}$ clearly admits always a global section, even if $\mathcal{G}$ is non-trivial; just take the section which assigns each base point the neutral element of its fibre.
\end{remarks}
If $M$ is a point we recover the notion of Lie groups, and, as usual, we have the notion of trivial LGBs:
\begin{examples}{Trivial examples}{TrivialLGBundle}
The \textbf{trivial LGB} is given as the product manifold $M \times G \to M$ with canonical multiplication $(x, g) \cdot (x, q) \coloneqq (x, gq)$, and we recover the notion of a Lie group in the case of $M = \{*\}$.
\end{examples}
We are of course also interested into LGB bundle morphisms:
\begin{definitions}{LGB morphism, \newline \cite[\S 1.2, special situation of Def.\ 1.2.1 \& 1.2.3, page 12]{mackenzieGeneralTheory}}{LGB morphism}
Let $\mathcal{G} \stackrel{\pi_{\mathcal{G}}}{\to} M$ and $\mathcal{H} \stackrel{\pi_{\mathcal{H}}}{\to} N$ be two LGBs over two smooth manifolds $M$ and $N$. An \textbf{LGB morphism} is a pair of smooth maps $F: \mathcal{H} \to \mathcal{G}$ and $f: N \to M$ such that
\ba\label{FibreRelationOverf}
\pi_{\mathcal{G}} \circ F &= f \circ \pi_{\mathcal{H}},\\
F(gq) &= F(g) ~ F(q)\label{LGBHomomorph}
\ea
for all $g, q \in \mathcal{H}$ with $\pi_{\mathcal{H}}(g) = \pi_{\mathcal{H}}(q)$. We then also say that \textbf{$F$ is an LGB morphism over $f$}. If $N = M$ and $f = \mathrm{id}_M$, then we often omit mentioning $f$ explicitly and just write that \textbf{$F$ is a (base-preserving) LGB morphism}.
We speak of an \textbf{LGB isomorphism (over $f$)} if $F$ is a diffeomorphism.
\end{definitions}
\begin{remark}\label{LGBMOrphismRemark}
\leavevmode\newline
\indent $\bullet$ The right hand side of Eq.\ \eqref{LGBHomomorph} is well-defined because of Eq.\ \eqref{FibreRelationOverf}.
$\bullet$ It is clear that condition 2 in Def.\ \ref{def:LieGroupBundle} is equivalent to say that $\mathcal{G}$ is locally isomorphic to a trivial LGB; as one may have expected already.
$\bullet$ If $F$ is a diffeomorphism, then also $f$: By Eq.\ \eqref{FibreRelationOverf} surjectivity of $f$ is clear; for $y \in M$ just take any $g \in \mathcal{G}_y$, and since $F$ is a bijective, we have a $q \in \mathcal{H}_x$ for some $x\in N$ with $F(q)=g$. By Eq.\ \eqref{FibreRelationOverf} we have $y = \pi_{\mathcal{G}}(F(q)) \stackrel{\eqref{FibreRelationOverf}}{=} f(x)$, thence, surjectivity follows. For injectivity we know by Eq.\ \eqref{LGBHomomorph} and \eqref{FibreRelationOverf} that $F\mleft(e^{\mathcal{H}}_x\mright) = e^{\mathcal{G}}_{f(x)}$, where $e^{\mathcal{H}}_x$ and $e^{\mathcal{G}}_{f(x)}$ denote the unique neutral elements of $\mathcal{H}_x$ and $\mathcal{G}_{f(x)}$, respectively. Assume that there are $x, x^\prime \in N$ with $f(x) = f(x^\prime)$, then we can derive
\bas
F\mleft(e^{\mathcal{H}}_x\mright)
&=
e^{\mathcal{G}}_{f(x)}
=
e^{\mathcal{G}}_{f(x^\prime)}
=
F\mleft(e^{\mathcal{H}}_{x^\prime}\mright).
\eas
Then we have $e^{\mathcal{H}}_x = e^{\mathcal{H}}_{x^\prime}$ due to that $F$ is bijective, and hence $x = x^\prime$. Therefore $f$ is bijective. Finally, $F^{-1}$ is by assumption also a diffeomorphism, Eq.\ \eqref{LGBHomomorph} clearly carries over, and Eq.\ \eqref{FibreRelationOverf} is clearly w.r.t.\ $f^{-1}$, that is
\bas
\pi_{\mathcal{H}} \circ F^{-1} &= f^{-1} \circ \pi_{\mathcal{G}}.
\eas
Since $\pi_{\mathcal{H}} \circ F^{-1}$ is smooth and $\pi_{\mathcal{G}}$ is a smooth surjective submersion, it follows that $f^{-1}$ is smooth; this is a well-known fact for right-compositions with surjective submersions, see \textit{e.g.}\ \cite[\S 3.7.2, Lemma 3.7.5, page 153]{Hamilton}. We can conclude that $f$ is a diffeomorphism. Observe that we also concluded that $F^{-1}$ is an LGB isomorphism, too.
\end{remark}
Similar to the case of Lie groups, \textit{the} example of an LGB are the automorphisms of a vector bundle.
\begin{examples}{Automorphisms of a vector bundle,\newline \cite[\S 1.1, special situation of Ex.\ 1.1.12, page 8]{mackenzieGeneralTheory}}{AutOfVectorBundleAnLGB}
Let $V\to M$ be a vector bundle and $\mathrm{Aut}(V) \to M$ its bundle of fibre-wise automorphisms (not to be confused with the sections of $\mathrm{Aut}(V)$ which are the base-preserving automorphisms of $V$). Denote with $W$ the structural vector space of $V$, then $\mathrm{Aut}(V)$ is an LGB with structural Lie group $\mathrm{Aut}(W)$. It is clear that each fibre of $\mathrm{Aut}(V)$ is a Lie group, and the LGB atlas is directly inherited by a vector bundle atlas $\mleft\{ \mleft( U_i, L_i \mright) \mright\}_{i \in I}$ of $V$, where we use a similar notation as for LGB atlases, especially we have vector bundle trivializations $L_{i}: \mleft.V\mright|_{U_i} \to U_i \times W$. Then define an LGB atlas over the same open covering $\mleft( U_i \mright)_i$ by
\bas
\mleft. \mathrm{Aut}(V) \mright|_{U_i} &\to U_i \times \mathrm{Aut}(W),\\
T &\mapsto \mleft.L_{i} \circ T \circ L_{i}^{-1}\mright|_{\{x\} \times W},
\eas
where $T \in \mleft.\mathrm{Aut}(V)\mright|_x = \mathrm{Aut}(V_x)$,
and $U_i \times \mathrm{Aut}(W)$ acts canonically on $U_i \times W$ in a fibre-wise sense. Then it is trivial to check that these give local trivializations such that $\mathrm{Aut}(V)$ carries the structure as an LGB.
\end{examples}
\subsection{Associated Lie group bundles}
For another important example recall that there is the notion of associated fibre bundles; following and stating the results of \cite[\S1, Construction 1.3.8, page 20]{mackenzieGeneralTheory} and \cite[\S 4.7, page 237ff.; see also Rem.\ 4.7.8, page 242f.]{Hamilton}: Let $P \stackrel{\pi_P}{\to} M$ be a principal bundle with structural Lie group $G$, a smooth manifold $N$ and a smooth left $G$-action $\Psi$ given by
\bas
G \times N &\to N,\\
(g, v) &\mapsto \Psi(g, v) \coloneqq g \cdot v.
\eas
Then we have a right $G$-action on $P \times N$ given by
\bas
(P \times N) \times G &\to P \times N,\\
(p,v,g) &\mapsto \mleft( p \cdot g, g^{-1} \cdot v \mright),
\eas
and one can show that the quotient under this action, $P\times_\Psi N \coloneqq ( P \times N) \Big/ G$, yields the structure of a fibre bundle
\begin{center}
\begin{tikzcd}
N \arrow{r} & P\times_\Psi N \arrow{d}{\pi_{P\times_\Psi N}} \\
& M
\end{tikzcd}
\end{center}
such that the projection $P \times N \to P \times_\Psi N$ is a smooth surjective submersion,
where the projection $\pi_{P\times_\Psi N}: P\times_\Psi N \to M$ is given by
\bas
\pi_{P\times_\Psi N}\mleft( [p, v] \mright)
&\coloneqq
\pi_P(p)
\eas
for all $[p, v] \in P\times_\Psi N$, denoting equivalence classes of $(p, v)$ by square brackets. For $x \in M$, the fibre $\mleft(P\times_\Psi N\mright)_x$ is given by $\mleft( P_x \times N \mright) \Big/ G = P_x \times_\Psi N$, and the fibre is diffeomorphic to $N$ by $N \ni v \mapsto [p, v] \in \mleft(P\times_\Psi N\mright)_x$ for a fixed $p \in P_x$. We will frequently use this diffeomorphism in the following without further notice.
A very important example are of course associated vector bundles, related to $N$ being a vetor space. We need a similar concept for Lie groups.
\begin{definitions}{Lie group representation on Lie groups, \newline \cite[special situation of the comment after Ex.\ 1.7.14, page 47]{mackenzieGeneralTheory}}{LieGroupActingOnLieGroup}
Let $G, H$ be Lie groups. Then a \textbf{Lie group representation of $G$ on $H$} is a smooth left action $\psi$ of $G$ on $H$
\bas
G \times H
&\to H,\\
(g,h)
&\mapsto
\psi_g(h)
\coloneqq
\psi(g, h)
\eas
such that
\ba
\psi_g(hq)
&=
\psi_g(h)
~ \psi_g(q)
\ea
for all $g \in G$ and $h,q \in H$.
\end{definitions}
\begin{remarks}{Note about labeling}{WhyRepresentation}
Observe that we have by the definition of group actions
\bas
\psi_{gg^\prime}
&=
\psi_g \circ \psi_{g^\prime}
\eas
for all $g, g^\prime \in G$, viewing $\psi_g$ as a map $H \to H$. Therefore we can view the action $\psi$ as a homomorphism
\bas
G &\to \mathrm{Aut}(H),
\eas
where $\mathrm{Aut}(H)$ is the set of Lie group automorphisms. The similarity to Lie group representations on vector spaces is obvious, thence the name.
\newline
This definition is of course also motivated by various references pointing out that Lie group representations define Lie group actions with extra properties; see for example \cite[\S 3, Ex.\ 3.4.2, page 143f.]{Hamilton}. In \cite[comments after Ex.\ 1.7.14, page 47]{mackenzieGeneralTheory} this definition is also called \textit{action by Lie group isomorphisms}.
\end{remarks}
With this we can discuss and define associated Lie group bundles; the following definition is clearly motivated by the definition of associated vector bundles as provided in \cite[\S 4, Thm.\ 4.7.2, page 239f.]{Hamilton}.
\begin{theorems}{Associated Lie group bundle as quotient}{AssociatedGroupBundlesHaveGroupStructure}
Let $G, H$ be Lie groups, $P \stackrel{\pi_P}{\to} M$ a principal $G$-bundle over a smooth manifold $M$, and $\psi$ a $G$-representation on $H$. Then $\mathcal{H} \coloneqq P \times_\psi H$ is an LGB
\begin{center}
\begin{tikzcd}
H \arrow{r} & \mathcal{H} \arrow{d}{\pi} \\
& M
\end{tikzcd}
\end{center}
with projection $\pi$ given by
\ba
\mathcal{H} &\to M,\nonumber\\
[p, h] &\mapsto \pi_P(p),
\ea
and fibres
\ba
\mathcal{H}_x
&=
P_x \times_\psi H
\ea
for all $x \in M$, which are isomorphic to $H$ as Lie groups. The Lie group structure on each fibre $\mathcal{H}_x$ is defined by
\ba\label{LiegroupStructureOnFibresofAssociated}
[p, h] \cdot \mleft[p, q\mright]
&\coloneqq
\mleft[ p, hq \mright]
\ea
for all $h, q \in H$ and $p_x \in P_x$, where $\pi_P(p) = x$.
\end{theorems}
\begin{remarks}{Neutral and inverse elements}{NeutralAndInverseInAssocLGB}
The neutral element for $\mathcal{H}_x$ ($x \in M$) is given by
\bas
e_x
&=
[p, e],
\eas
where $p \in P_x$ is arbitrary and $e$ is the neutral element of $H$. This is clearly independent of the choice of $p$ due to
\bas
\mleft[ p, e \mright]
&=
\mleft[ p \cdot g, \psi_{g^{-1}}(e) \mright]
=
\mleft[ p \cdot g, e \mright]
\eas
for all $g \in G$. Thence, the fact that $e_x$ is the neutral element follows immediately.
By Def.\ \eqref{LiegroupStructureOnFibresofAssociated} the inverse of $[p, h] \in \mathcal{H}_x$ is clearly given by
\bas
\mleft( [p, h] \mright)^{-1}
&=
\mleft[ p, h^{-1} \mright].
\eas
\end{remarks}
\begin{proof}
\leavevmode\newline
\indent $\bullet$ That $\pi$ is the well-defined projection and that the fibres are precisely $P_x \times_\psi H$ for all $x \in M$ is well-known, see our discussion before Def.\ \ref{def:LieGroupActingOnLieGroup} and the references therein; it is also very straightforward to check. We also discussed that $\mathcal{H}$ is a fibre bundle with structural fibre $H$. Hence, if one knows that the proposed group structure in Def.\ \eqref{LiegroupStructureOnFibresofAssociated} is well-defined, then the smoothness of the group structure is implied by the smoothness structures of $H$ and $\mathcal{H}$. Thence, let us check whether Def.\ \eqref{LiegroupStructureOnFibresofAssociated} is well-defined. Let $x \in M$, $p \in P_x$ and $p^\prime \coloneqq p \cdot g^\prime$ be another element of $P_x$, where $g^\prime \in G$. Also let $[p_1,h_1], [p_2, h_2] \in P_x \times_\psi H$; then we have unique elements $q_i, q_i^\prime$ of $G$ such that ($i \in \{1,2\}$)
\bas
p_i &= p \cdot q_i,&
p_i &= p^\prime \cdot q_i^\prime,
\eas
especially, it follows $q_i = g^\prime q_i^\prime$.
On the one hand, if we use $p$ as fixed element of $P_x$ to calculate the multiplication, we get
\ba\label{MultiPlicationInAssocGroup}
[p_1,h_1] \cdot [p_2,h_2]
&=
\mleft[ p, \psi_{q_1}(h_1) \mright]
\cdot \mleft[ p, \psi_{q_2}(h_2) \mright]
=
\mleft[ p, \psi_{q_1}(h_1) ~ \psi_{q_2}(h_2) \mright],
\ea
on the other hand, using Def.\ \ref{def:LieGroupActingOnLieGroup} and $p^\prime = p \cdot g^\prime$ instead of $p$,
\bas
[p_1,h_1] \cdot [p_2,h_2]
&=
\mleft[ p \cdot g^\prime, \psi_{q_1^\prime}(h_1) ~ \psi_{q_2^\prime}(h_2) \mright]
\\
&=
\Bigl[ p, \underbrace{\psi_{g^\prime} \mleft( \psi_{q_1^\prime}(h_1) ~ \psi_{q_2^\prime}(h_2) \mright)}_{= \psi_{g^\prime} \mleft( \psi_{q_1^\prime}(h_1) \mright) ~ \psi_{g^\prime} \mleft( \psi_{q_2^\prime}(h_2) \mright)} \Bigr]
\\
&=
\mleft[ p, \psi_{g^\prime q_1^\prime}(h_1) ~ \psi_{g^\prime q_2^\prime}(h_2) \mright]
\\
&=
\mleft[ p, \psi_{q_1}(h_1) ~ \psi_{q_2}(h_2) \mright],
\eas
which implies that Def.\ \eqref{LiegroupStructureOnFibresofAssociated} is well-defined, and thus defines a Lie group structure on each fibre of $\mathcal{H}$.
$\bullet$ That the fibres $\mathcal{H}_x$ are isomorphic to $H$ as Lie groups for all $x \in M$ also quickly follows. Recall by our discussion before Def.\ \ref{def:LieGroupActingOnLieGroup} that the fibres are diffeormorphic to $H$ by $H \ni h \mapsto [p, h] \in \mathcal{H}_x$ for a fixed $p \in P_x$. By Def.\ \eqref{LiegroupStructureOnFibresofAssociated} it is clear that this map is a Lie group homomorphism and hence a Lie group isomorphism.
$\bullet$ Let us now construct an LGB atlas for $\mathcal{H}$ by using a principal bundle atlas for $P$. That is, for some $U \subset M$ open and a trivialization $\varphi_U: P|_U \to U \times G$ we write
\bas
\varphi_U(p)
&=
\bigl( \pi_P(p), \beta_U(p) \bigr)
\eas
for all $p \in P$, where $\beta_U: P|_U \to G$ is an equivariant map, \textit{i.e.}\ $\beta_U(p \cdot g) = \beta_U(p) ~ g$ for all $g \in G$. Then define $\phi_U$ as a map by
\bas
\mathcal{H}|_U
&\to
U \times H,\\
[p, h]
&\mapsto
\mleft(
\pi_P(p), \psi_{\beta_U(p)} (h)
\mright).
\eas
$\phi_U$ is well-defined: Let $\mleft[p^\prime, h^\prime\mright] \in \mathcal{H}|_U$ with $\mleft[p^\prime, h^\prime\mright] = \mleft[p, h\mright]$. Then there is a $g \in G$ such that
\bas
\mleft(p^\prime, h^\prime\mright)
&=
\mleft( p \cdot g, \psi_{g^{-1}}(h) \mright),
\eas
hence, using the equivariance of $\beta_U$ and Def.\ \ref{def:LieGroupActingOnLieGroup},
\bas
\phi_U\mleft( \mleft[p^\prime, h^\prime\mright] \mright)
&=
\Bigl(
\underbrace{\pi_P\mleft(p \cdot g\mright)}_{= \pi_P(p)}, \underbrace{\mleft(\psi_{\beta_U\mleft(p \cdot g\mright)} \circ \psi_{g^{-1}} \mright)}_{= \psi_{\beta_U(p)} \circ \psi_g \circ \psi_{g^{-1}} } (h)
\Bigr)
=
\mleft(
\pi_P(p), \psi_{\beta_U(p)} (h)
\mright)
=
\phi_U\bigl( [p, h] \bigr),
\eas
which proves that $\phi_U$ is well-defined. Denote the projection onto equivalence classes $P \times H \to \mathcal{H}$ by $\varpi$, then observe
\bas
\phi_U \circ \varpi
&=
L,
\eas
where $L_U: P|_U \times H \to U \times H$ is given by $L_U(p,h) \coloneqq \mleft( \pi_P(p), \psi_{\beta_U(p)} (h) \mright)$ for all $(p, h) \in P|_U \times H$. $L_U$ is clearly smooth and recall that $\varpi$ is a smooth surjective submersion, therefore $\phi_U$ is smooth; this is a well-known fact for right-compositions with surjective submersions, see \textit{e.g.}\ \cite[\S 3.7.2, Lemma 3.7.5, page 153]{Hamilton}. We define a candidate of the inverse $\phi_U^{-1}: U \times H \to \mathcal{H}|_U$ by
\bas
\phi_U^{-1}(x, h)
&=
\mleft[ \varphi_U^{-1}\mleft(x, e\mright), h \mright]
\eas
for all $(x, h) \in U \times H$, where $e$ is the neutral element of $G$.
By the definition of $\varphi_U$ we immediately get
\bas
\mleft( \varphi_U \circ \varphi^{-1}_U \mright)(x, e)
&=
\Bigl(
\pi_P\mleft( \varphi_U^{-1}(x, e) \mright), \beta_U\mleft( \varphi_U^{-1}(x, e) \mright)
\Bigr)
=
(x, e),
\eas
for all $x \in U$, and, also using again the equivariance of $\beta_U$,
\bas
\varphi^{-1}_U\mleft(\pi_P(p), e\mright)
&=
\varphi^{-1}_U\Bigl(\pi_P\mleft(p \cdot \beta_U^{-1}(p) \mright), \beta_U(p)~\beta_U^{-1}(p)\Bigr)
\\
&=
\varphi^{-1}_U\Bigl(\pi_P\mleft(p \cdot \beta_U^{-1}(p) \mright), \beta_U\mleft(p \cdot \beta_U^{-1}(p)\mright)\Bigr)
\\
&=
\mleft(\varphi^{-1}_U \circ \varphi_U\mright)\mleft( p \cdot \beta_U^{-1}(p) \mright)
\\
&=
p \cdot \beta_U^{-1}(p)
\eas
for all $p \in P|_U$.
Then
\bas
\mleft(\phi_U \circ \phi_U^{-1}\mright)(x, h)
&=
\mleft(
\pi_P\mleft( \varphi^{-1}_U(x, e) \mright), \psi_{\beta_U\mleft( \varphi^{-1}_U(x, e) \mright)} (h)
\mright)
=
\bigl(
x, \psi_e(h)
\bigr)
=
(x, h),
\eas
for all $(x, h) \in U \times H$, and
\bas
\mleft(\phi_U^{-1} \circ \phi_U\mright)([p, h])
&=
\bigl[
\underbrace{\varphi_U^{-1}\mleft( \pi_P(p), e \mright)}_{= p \cdot \beta_U^{-1}(p) },
\psi_{\beta_U(p)}(h)
\bigr]
\\
&=
\mleft[