Repository navigation
Expand file tree
/
Copy pathT1S1_Basic_Definition.tex
More file actions
735 lines (676 loc) · 31.3 KB
/
Copy pathT1S1_Basic_Definition.tex
File metadata and controls
735 lines (676 loc) · 31.3 KB
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
\section{Basic Definition}
\subsection{Action on the extended complex plane}\label{Sec:MF_action}
$\GL_2(\bbC)$ acts on $\widehat{\bbC}$. For $g = \begin{pmatrix}a & b\\ c & d\end{pmatrix}$ and $z \in \widehat{\bbC}$,
\[
gz \defeq g \cdot z = \begin{cases}
\dfrac{az + b}{cz + d}, & z \in \bbC, \ cz + d \ne 0, \\[5pt]
\infty, & z \in \bbC, \ cz + d = 0, \\
\infty, & z = \infty, \ c = 0, \\
\dfrac ac, & z = \infty, \ c \ne 0.
\end{cases}
\]
\[
Z(\GL_2(\bbC)) = \left\{\begin{pmatrix}a & 0\\ 0 & a\end{pmatrix} : a \in \bbC^\times\right\}.
\]
\[
\calH = \{z \in \bbC : \Im(z) > 0\}, \qquad\text{upper half-plane}.
\]
\begin{proposition}\label{Prop:MF_imaginary_part}
Let $g \in \GL_2(\bbR)$ and $z \in \bbC$. If $cz + d \ne 0$, then
\[
\Im(gz) = \frac{\det(g)}{|cz + d|^2}\Im(z).
\]
\end{proposition}
\begin{proof}
\begin{align*}
\Im(gz)
&= \Im\left(\frac{(az + b)(c\bar z + d)}{(cz + d)(c\bar z + d)}\right)\\
&= \Im\left(\frac{ac z\bar z + bd + ad z + bc\bar z}{|cz + d|^2}\right)\\
&= \frac{ad\Im(z) - bc\Im(z)}{|cz + d|^2}\\
&= \frac{(ad - bc)\Im(z)}{|cz + d|^2}.
\end{align*}
\end{proof}
\begin{lemma}\label{Lem:MF_denominator}
If $g = \begin{pmatrix}a & b\\ c & d\end{pmatrix} \in \GL_2(\bbR)$ and $z \in \bbC \setminus \bbR$, then $cz + d \ne 0$.
\end{lemma}
because $\det(g) \ne 0 \Longrightarrow (c, d) \ne (0, 0)$.
\begin{corollary}\label{Cor:MF_half_plane}
The action of $\GL_2^+(\bbR)$ on $\widehat{\bbC}$ stabilizes $\calH$.
\end{corollary}
\subsection{Congruence Subgroups}\label{Sec:MF_congruence}
For $N \in \bbZ_{\ge 1}$, let
\[
\Gamma(N) = \left\{g = \begin{pmatrix}a & b\\ c & d\end{pmatrix} \in \SL_2(\bbZ) :
\begin{pmatrix}a & b\\ c & d\end{pmatrix} \equiv
\begin{pmatrix}1 & 0\\ 0 & 1\end{pmatrix}\pmod N\right\}.
\]
This is the \termino{principal congruence subgroup of level $N$}.
\begin{proposition}\label{Prop:MF_principal_index}
Let $N \ge 1$. Then:
\begin{enumerate}[label=(\arabic*)]
\item The reduction map
\[
\pi_N : \SL_2(\bbZ) \longrightarrow \SL_2(\bbZ/N\bbZ)
\]
is surjective, with kernel $\Gamma(N)$. In particular,
\[
\SL_2(\bbZ)/\Gamma(N) \cong \SL_2(\bbZ/N\bbZ).
\]
\item The index of $\Gamma(N)$ in $\SL_2(\bbZ)$ is
\[
[\SL_2(\bbZ) : \Gamma(N)]
= N^3\prod_{p \mid N}\left(1 - \frac{1}{p^2}\right).
\]
\end{enumerate}
\end{proposition}
\begin{proof}
\leavevmode
\begin{enumerate}[label=(\arabic*)]
\item Let $\gamma \in \SL_2(\bbZ/N\bbZ)$. Choose an integral lift
\[
A = \begin{pmatrix}a & b\\ c & d\end{pmatrix} \in M_2(\bbZ),
\qquad A \equiv \gamma\pmod N.
\]
Consider the set of all lifts
\[
S = \left\{A + N\begin{pmatrix}k & \ell\\ s & t\end{pmatrix} :
k, \ell, s, t \in \bbZ\right\}.
\]
We want to prove $S \cap \SL_2(\bbZ) \ne \varnothing$.
\begin{enumerate}[label=\textit{Step~\arabic*.}]
\item Find $c'$ and $d'$ with $\gcd(c', d') = 1$.
We seek integers $s, t$ such that
\[
c' = c + sN, \qquad d' = d + tN, \qquad \gcd(c', d') = 1.
\]
Since $\det\gamma = 1$ in $\bbZ/N\bbZ$, we have
\[
ad - bc \equiv 1\pmod N.
\]
Set $g = \gcd(c, d)$. As $g \mid ad - bc$, this implies
\[
\boxed{\gcd(g, N) = 1.}
\]
(The case $N = 1$ is immediate, so we may assume $N > 1$.)
The argument is symmetric in $c$ and $d$. Moreover, $c$ and $d$ cannot both be zero, since $ad - bc \equiv 1\pmod N$. Thus, without loss of generality, assume $c \ne 0$ and take $s = 0$. In other words, put $c' = c$ and look for
\[
d' = d + tN\qquad\text{with}\qquad\gcd(c, d + tN) = 1.
\]
By the Chinese remainder theorem, there exists $t \in \bbZ$ such that
\[
\boxed{\begin{cases}
t \equiv 1\pmod p, & \text{for every prime }p \mid g, \\
t \equiv 0\pmod p, & \text{for every prime }p \mid c\text{ with }p \nmid g.
\end{cases}}
\]
We now check that $\gcd(c, d + tN) = 1$. Let $p$ be any prime divisor of $c$.
\begin{itemize}[leftmargin=2em, nosep]
\item If $p \mid g$, then $p \mid d$ but $p \nmid N$, since $\gcd(g, N) = 1$. Hence
\[
d + tN \equiv N\not \equiv 0\pmod p.
\]
\item If $p \nmid g$, then $p \nmid d$, and the choice $t \equiv 0\pmod p$ gives
\[
d + tN \equiv d\not \equiv 0\pmod p.
\]
\end{itemize}
Thus no prime divisor of $c$ divides $d + tN$, proving
\[
\boxed{\gcd(c', d') = \gcd(c, d + tN) = 1.}
\]
\item Find $k, \ell \in \bbZ$ to make the determinant equal to $1$.
We want to find integers $k, \ell$ such that
\[
\det\begin{pmatrix}a + kN & b + \ell N\\ c' & d'\end{pmatrix} = 1.
\]
Since $c' \equiv c$ and $d' \equiv d\pmod N$ and $ad - bc \equiv 1\pmod N$, we have $ad' - bc' \equiv 1\pmod N$. Thus
\[
m \defeq \frac{1 - (ad' - bc')}{N} \in \bbZ.
\]
Expanding the determinant, the required equation becomes
\[
(ad' - bc') + N(kd' - \ell c') = 1,
\qquad\text{equivalently}\qquad kd' - \ell c' = m.
\]
By Step 1, $\gcd(c', d') = 1$. Hence B\'ezout's identity gives $u, v \in \bbZ$ such that $ud' + vc' = 1$. Taking $k = mu$ and $\ell = - mv$ gives $kd' - \ell c' = m$, as required. Therefore
\[
\begin{pmatrix}a + kN & b + \ell N\\ c' & d'\end{pmatrix}
\in S \cap \SL_2(\bbZ).
\]
\end{enumerate}
Consequently $S \cap \SL_2(\bbZ) \ne \varnothing$, so $\pi_N$ is surjective. Its kernel is $\Gamma(N)$ by definition; the first isomorphism theorem gives
\[
\SL_2(\bbZ)/\Gamma(N) \cong \SL_2(\bbZ/N\bbZ).
\]
\item By (1),
\[
[\SL_2(\bbZ) : \Gamma(N)] = \#\SL_2(\bbZ/N\bbZ).
\]
Write $N = \prod_{p \mid N}p^{n_p}$. The Chinese remainder theorem gives an isomorphism of rings
\[
\bbZ/N\bbZ \cong \prod_{p \mid N}\bbZ/p^{n_p}\bbZ.
\]
which induces a group isomorphism
\[
\SL_2(\bbZ/N\bbZ) \cong
\prod_{p \mid N}\SL_2(\bbZ/p^{n_p}\bbZ).
\]
It therefore suffices to compute $\#\SL_2(\bbZ/p^e\bbZ)$ for a prime $p$ and $e \ge 1$.
\begin{enumerate}[label=\textbf{Case~\arabic*}]
\item $e > 1$.
For $e \ge 2$, consider the reduction homomorphism
\[
\lambda_e : \SL_2(\bbZ/p^e\bbZ)
\longrightarrow \SL_2(\bbZ/p^{e - 1}\bbZ).
\]
This map is surjective by (1): the surjective reduction $\pi_{p^{e - 1}}$ factors as $\lambda_e \circ \pi_{p^e}$. An element lies in $\ker\lambda_e$ precisely when it has the form
\[
I + p^{e - 1}A\pmod{p^e}, \qquad A \in M_2(\bbF_p),
\]
where any integral lift of $A$ may be used. As $2(e - 1) \ge e$,
\[
\det(I + p^{e - 1}A)
\equiv 1 + p^{e - 1}\operatorname{tr}(A)\pmod{p^e}.
\]
Thus the determinant is $1$ modulo $p^e$ exactly when $\operatorname{tr}(A) = 0$ in $\bbF_p$. Since three entries of $A$ can be chosen freely and the fourth is fixed by the trace condition,
\[
\#\ker\lambda_e = p^3.
\]
It follows that
\[
\#\SL_2(\bbZ/p^e\bbZ)
= p^3\#\SL_2(\bbZ/p^{e - 1}\bbZ)
= p^{3(e - 1)}\#\SL_2(\bbF_p).
\]
\item $e = 1$.
We use the determinant homomorphism:
\[
1 \longrightarrow \SL_2(\bbF_p)
\longrightarrow \GL_2(\bbF_p)
\xrightarrow{\ \det\ }\bbF_p^\times
\longrightarrow 1.
\]
It is surjective. To count $\GL_2(\bbF_p)$, choose a nonzero first column ($p^2 - 1$ choices) and then a second column outside its span ($p^2 - p$ choices). Thus
\[
\#\GL_2(\bbF_p) = (p^2 - 1)(p^2 - p),
\qquad \#\bbF_p^\times = p - 1.
\]
Hence
\[
\#\SL_2(\bbF_p)
= \frac{(p^2 - 1)(p^2 - p)}{p - 1}
= p(p^2 - 1)
= p^3\left(1 - \frac1{p^2}\right).
\]
\end{enumerate}
Consequently,
\[
\#\SL_2(\bbZ/p^e\bbZ)
= p^{3e}\left(1 - \frac1{p^2}\right).
\]
Taking the product over all primes dividing $N$ proves
\[
[\SL_2(\bbZ) : \Gamma(N)]
= \prod_{p \mid N}p^{3n_p}\left(1 - \frac1{p^2}\right)
= \boxed{N^3\prod_{p \mid N}\left(1 - \frac1{p^2}\right)}.
\]
\end{enumerate}
\end{proof}
\begin{definition}\label{Def:MF_congruence}
A subgroup $\Gamma$ of $\SL_2(\bbZ)$ is a \termino{congruence subgroup} if $\Gamma(N) \subseteq \Gamma$ for some $N$.
\end{definition}
\begin{example}
\begin{align*}
\Gamma_1(N) &= \left\{g = \begin{pmatrix}a & b\\ c & d\end{pmatrix} \in \SL_2(\bbZ) :
\begin{pmatrix}a & b\\ c & d\end{pmatrix} \equiv
\begin{pmatrix}1 & *\\ 0 & 1\end{pmatrix}\pmod N\right\}, \\
\Gamma_0(N) &= \left\{g = \begin{pmatrix}a & b\\ c & d\end{pmatrix} \in \SL_2(\bbZ) :
\begin{pmatrix}a & b\\ c & d\end{pmatrix} \equiv
\begin{pmatrix}* & *\\ 0 & *\end{pmatrix}\pmod N\right\}.
\end{align*}
Here $*$ means no condition on the entry.
\[
\Gamma(N) \subseteq \Gamma_1(N) \subseteq \Gamma_0(N) \subseteq \SL_2(\bbZ).
\]
\end{example}
\begin{proposition}\label{Prop:MF_congruence_quotients}
Let $N \in \bbN_{\ge 1}$.
\begin{enumerate}[label=\arabic*)]
\item The map
\[
\lambda_1 : \Gamma_1(N) \longrightarrow \bbZ/N\bbZ,
\qquad\begin{pmatrix}a & b\\ c & d\end{pmatrix} \longmapsto b\pmod N
\]
is a surjective group homomorphism with kernel $\Gamma(N)$.
\item The map
\[
\lambda_0 : \Gamma_0(N) \longrightarrow (\bbZ/N\bbZ)^\times,
\qquad\begin{pmatrix}a & b\\ c & d\end{pmatrix} \longmapsto d\pmod N
\]
is a surjective group homomorphism with kernel $\Gamma_1(N)$.
\end{enumerate}
\end{proposition}
\noindent Indeed,
\[
\lambda_1(gg') = \lambda_1(g) + \lambda_1(g')\pmod N.
\]
To see that $\lambda_0$ is well-defined, note that
\[
1 = \det(g) = ad - bc \equiv ad\pmod N
\quad \Longrightarrow \quad d \in (\bbZ/N\bbZ)^\times.
\]
Also,
\[
\lambda_0(gg') = \lambda_0(g)\lambda_0(g').
\]
\subsubsection*{Summary} $\SL_2(\bbZ)$ acts on $\calH$.
\[
0 \longrightarrow \Gamma(N) \longrightarrow \SL_2(\bbZ)
\longrightarrow \SL_2(\bbZ/N\bbZ) \longrightarrow 0,
\]
\[
0 \longrightarrow \Gamma(N) \longrightarrow \Gamma_1(N)
\longrightarrow \bbZ/N\bbZ \longrightarrow 0,
\]
\[
0 \longrightarrow \Gamma_1(N) \longrightarrow \Gamma_0(N)
\longrightarrow (\bbZ/N\bbZ)^\times\longrightarrow 0.
\]
\[
\begin{pmatrix}a & b\\ 0 & d\end{pmatrix}
\begin{pmatrix}1 & x\\ 0 & 1\end{pmatrix}
= \begin{pmatrix}a & ax + b\\ 0 & d\end{pmatrix}.
\]
\[
\begin{pmatrix}a & b\\ 0 & d\end{pmatrix}
\begin{pmatrix}1 & -d^{-1}b\\ 0 & 1\end{pmatrix}
= \begin{pmatrix}a & 0\\ 0 & d\end{pmatrix}.
\]
\subsection{Weak Modular Functions}\label{Sec:MF_weak}
\begin{definition}\label{Def:MF_automorphy}
For $g = \begin{pmatrix}a & b\\ c & d\end{pmatrix} \in \SL_2(\bbZ)$ and $z \in \calH$, the \termino{factor of automorphy} is
\[
j(g, z) = cz + d \in \bbC^\times.
\]
\end{definition}
\begin{definition}\label{Def:MF_slash}
For $g \in \SL_2(\bbZ)$ and $k \in \bbZ$, the weight-$k$ operator $[g]_k$ associates to a function $f : \calH \to \bbC$ the function $f[g]_k : \calH \to \bbC$ defined by
\[
(f[g]_k)(z) = j(g, z)^{-k}f(gz), \qquad z \in \calH.
\]
\end{definition}
\begin{lemma}\label{Lem:MF_slash}
Let $g, g' \in \SL_2(\bbZ)$, $z \in \calH$, and $f : \calH \to \bbC$. For every $k \in \bbZ$, the following identities hold:
\begin{enumerate}[label=(\arabic*)]
\item $j(gg', z) = j(g, g'z)\,j(g', z)$.
\item $f[gg']_k = (f[g]_k)[g']_k$.
\item $\displaystyle\frac{d}{dz}(gz) = \frac{1}{j(g, z)^2}$.
\end{enumerate}
\end{lemma}
These three identities can be verified by direct calculation and are left as an exercise.
\begin{definition}\label{Def:MF_weak}
Let $\Gamma$ be a congruence subgroup and let $k \in \bbZ$. A function $f : \calH \to \bbC$ is \termino{weakly modular of weight $k$} if
\begin{enumerate}[label=(\arabic*)]
\item $f$ is meromorphic;
\item $f[\gamma]_k = f$ for every $\gamma \in \Gamma$.
\end{enumerate}
\end{definition}
\begin{remarkstar}
By \lemref{Lem:MF_denominator}, $j(g, z)$ is nonzero on the upper half-plane. Hence if $f$ is meromorphic, then $f[g]_k$ is also meromorphic.
\end{remarkstar}
\begin{remark}
When $k = 0$, the weak modularity condition is $f(\gamma z) = f(z)$ for every $\gamma \in \Gamma$. In other words, $f$ is a $\Gamma$-invariant meromorphic function. In particular, constant functions are weakly modular of weight $0$.
\end{remark}
\begin{remark}
If $ - I = \begin{pmatrix} - 1 & 0\\ 0 & - 1\end{pmatrix} \in \Gamma$ and $f$ is weakly modular of weight $k$ with respect to $\Gamma$, then
\[
f(z) = (f[-I]_k)(z) = (-1)^{-k}f(z).
\]
Thus if $k$ is odd, then $f \equiv 0$.
\end{remark}
\begin{remark}\label{Rmk:MF_period}
By the definition of a congruence subgroup, there exists $N \geqslant 1$ such that $\Gamma(N) \subseteq \Gamma$. In particular,
\[
\begin{pmatrix}1 & N\\ 0 & 1\end{pmatrix} \in \Gamma.
\]
Hence there exists a least positive integer $h$ such that
\[
\gamma_h = \begin{pmatrix}1 & h\\ 0 & 1\end{pmatrix} \in \Gamma.
\]
If $f$ is weakly modular of weight $k$ with respect to $\Gamma$, then
\[
f(z) = (f[\gamma_h]_k)(z) = (0z + 1)^{-k}f(\gamma_hz) = f(z + h).
\]
Thus $f$ is $h\bbZ$-periodic. We will use this when introducing $q$-expansions.
\end{remark}
\begin{remark}
Suppose that the congruence subgroup $\Gamma = \langle\gamma_1, \cdots, \gamma_n\rangle$.
By \lemref{Lem:MF_slash}(2), to check weak modularity for a meromorphic function $f$, it suffices to verify
\[
f[\gamma_i]_k = f, \qquad i = 1, \cdots, n,
\]
These conditions imply that $f$ is weakly modular of weight $k$ with respect to $\Gamma$. Indeed, every element of $\Gamma$ is a product of generators and their inverses, and the composition law shows that invariance under the generators implies invariance under their inverses and all such products.
\end{remark}
\begin{example}
$\SL_2(\bbZ)$ is generated by the following two matrices:
\[
S = \begin{pmatrix}0 & - 1\\ 1 & 0\end{pmatrix}, \qquad
T = \begin{pmatrix}1 & 1\\ 0 & 1\end{pmatrix}.
\]
Hence a meromorphic function $f : \calH \to \bbC$ satisfying $f[S]_k = f$ and $f[T]_k = f$ is weakly modular of weight $k$ with respect to $\SL_2(\bbZ)$. Explicitly, these two conditions are
\[
f(z) = z^{-k}f(-1/z), \qquad f(z + 1) = f(z)(\bbZ\text{-periodic}).
\]
\end{example}
\begin{lemma}\label{Lem:MF_conjugation}
\leavevmode
\begin{enumerate}[label=(\roman*)]
\item If $\Gamma$ is a congruence subgroup and $\alpha \in \SL_2(\bbZ)$, then $\alpha^{-1}\Gamma\alpha$ is a congruence subgroup.
\item If $f$ is weakly modular of weight $k$ with respect to $\Gamma$ and $\alpha \in \SL_2(\bbZ)$, then $f[\alpha]_k$ is weakly modular of weight $k$ with respect to $\alpha^{-1}\Gamma\alpha$.
\end{enumerate}
\end{lemma}
\begin{proof}
\leavevmode
\begin{enumerate}[label=(\roman*)]
\item Let $\Gamma(N) \subseteq \Gamma$. Since $\Gamma(N) \trianglelefteq \SL_2(\bbZ)$,
\[
\Gamma(N) = \alpha^{-1}\Gamma(N)\alpha \subseteq \alpha^{-1}\Gamma\alpha.
\]
\item By definition and \lemref{Lem:MF_slash}(2),
\[
(f[\alpha]_k)[\alpha^{-1}\gamma\alpha]_k
= f[\alpha\alpha^{-1}\gamma\alpha]_k
= f[\gamma\alpha]_k
= (f[\gamma]_k)[\alpha]_k = f[\alpha]_k
\]
for $\gamma \in \Gamma$.
\end{enumerate}
\end{proof}
\subsection{Cusps}\label{Sec:MF_cusps}
Put $X = \bbQ \cup \{\infty\} = \bbP^1(\bbQ)$. Since $\SL_2(\bbZ)$ stabilizes $X$, its action on $X$ is well-defined. We call the $\Gamma$-orbits in $X$ the \termino{cusps} of $\Gamma$.
Let $\Gamma$ be a congruence subgroup. For $x \in X$, its \termino{$\Gamma$-orbit} is
\[
\Gamma \cdot x = \{\gamma x : \gamma \in \Gamma\}.
\]
We write $x \sim _{\Gamma}y$ if $y = \gamma x$ for some $\gamma \in \Gamma$, and denote the corresponding orbit (equivalence class) by $[x]_{\Gamma} = \Gamma \cdot x$.
\begin{definition}\label{Def:MF_cusps}
The \termino{set of cusps} of $\Gamma$ is the orbit set
\[
\Gamma \backslash X
= \Gamma \backslash \bigl(\bbQ \cup \{\infty\}\bigr)
= \{[x]_{\Gamma} : x \in X\}.
\]
\end{definition}
\begin{lemma}\label{Lem:MF_cusps}
\leavevmode
\begin{enumerate}[label=(\roman*)]
\item $\SL_2(\bbZ) \backslash X = \{[\infty]_{\SL_2(\bbZ)}\}$.
\item The set of cusps $\Gamma \backslash X$ is finite.
\end{enumerate}
\end{lemma}
\begin{proof}
\leavevmode
\begin{enumerate}[label=(\roman*)]
\item Let $x = a/c \in \bbQ$, with $a, c \in \bbZ$ and $\gcd(a, c) = 1$. By B\'ezout's identity there are $b, d \in \bbZ$ such that $ad - bc = 1$. Hence
\[
\gamma = \begin{pmatrix}a & b\\ c & d\end{pmatrix}
\in \SL_2(\bbZ), \qquad \gamma\infty = \frac ac = x.
\]
The point $\infty$ itself is in its own orbit, so every element of $X$ is in the orbit of $\infty$.
\item Put $G = \SL_2(\bbZ)$. Since $\Gamma$ is a
congruence subgroup, there exists $N \ge 1$ with $H \defeq \Gamma(N) \subseteq \Gamma$. By \propref{Prop:MF_principal_index}, $H \trianglelefteq G$ and $G/H$ is finite. Write $\overline{g} = gH \in G/H$ and $[x]_H = H \cdot x \in H \backslash X$. The group $G/H$ acts on the $H$-orbit space $H \backslash X$ via
\[
\overline{g} \cdot [x]_H \defeq [gx]_H.
\]
This is well-defined because $H$ is normal in $G$. Taking orbits in two stages gives
\[
(G/H) \backslash (H \backslash X) \cong G \backslash X
= \{[\infty]_G\}.
\]
Thus the finite group $G/H$ acts transitively on $H \backslash X$; in particular,
\[
|H \backslash X| \le |G/H| < \infty.
\]
Since $H \subseteq \Gamma$, there is a natural surjection $H \backslash X \twoheadrightarrow \Gamma \backslash X$, so
\[
|\Gamma \backslash X| < \infty.
\]
\end{enumerate}
\end{proof}
Let
\[
D = \{q \in \bbC : |q| < 1\}, \qquad D' = D \setminus \{0\}
\]
be the \termino{open complex unit disk} and the \termino{punctured open complex unit disk}, respectively. Let $h \in \bbN_{ > 0}$. The function
\[
q_h : \calH \longrightarrow D', \qquad
z \longmapsto q_h(z) = e^{2\pi iz/h}
\]
is $h\bbZ$-periodic.
Let $f : \calH \to \bbC$ be an $h\bbZ$-periodic function. Since $q_h$ identifies $\calH/h\bbZ$ with $D'$, the function $f$ factors through $q_h$: there is a function $\widetilde f : D' \to \bbC$ such that
\[
f(z) = \widetilde f(q_h(z)).
\]
\begin{definition}\label{Def:MF_at_infinity}
We say that $f$ is \termino{meromorphic} (respectively, \termino{holomorphic}) \termino{at $\infty$} if $\widetilde f$ extends meromorphically (respectively, holomorphically) across $0 \in D$.
\end{definition}
If $f$ is meromorphic at $\infty$, then $\widetilde f$ has a Laurent expansion at $0$. Thus $f$ has a \termino{Fourier expansion} (or \termino{$q_h$-expansion})
\[
\boxed{f(z) = \sum_{n = - m}^{\infty}a(n)\,q_h(z)^n, \qquad m \ge 0, }
\]
valid for $\Im(z)$ sufficiently large. In the holomorphic case there are no negative powers (we can take $m = 0$), and
\[
f(\infty) \defeq \widetilde f(0) = a(0).
\]
\subsection{Modular Forms}\label{Sec:MF_forms}
\begin{definition}\label{Def:MF_modular_form}
Let $k \in \bbZ$ and let $\Gamma$ be a congruence subgroup. A \termino{modular form of weight $k$} with respect to $\Gamma$ is a function $f : \calH \to \bbC$ such that
\begin{enumerate}[label=(\arabic*)]
\item $f$ is holomorphic on $\calH$;
\item $f$ is weakly modular of weight $k$ with respect to $\Gamma$;
\item for every $\alpha \in \SL_2(\bbZ)$, the function $f[\alpha]_k$ is holomorphic at $\infty$.
\end{enumerate}
\end{definition}
\begin{definition}\label{Def:MF_cusp_form}
A modular form of weight $k$ with respect to $\Gamma$ is a \termino{cusp form} if
\[
\forall\alpha \in \SL_2(\bbZ), \qquad (f[\alpha]_k)(\infty) = 0.
\]
\end{definition}
\begin{remark}
Condition (3) of \defref{Def:MF_modular_form} only needs to be checked for finitely many $\alpha$. Indeed, the congruence subgroup $\Gamma$ has finite index. Choose representatives $\alpha_1, \cdots, \alpha_r$ of the left-coset space $\Gamma \backslash \SL_2(\bbZ)$. Every $\alpha \in \SL_2(\bbZ)$ can be written as $\alpha = \gamma\alpha_i$ with $\gamma \in \Gamma$. By \lemref{Lem:MF_slash}(2),
\[
f[\alpha]_k = f[\gamma\alpha_i]_k = (f[\gamma]_k)[\alpha_i]_k = f[\alpha_i]_k.
\]
Thus it suffices to check that each $f[\alpha_i]_k$ is holomorphic at $\infty$.
\end{remark}
\begin{example}
Equipped with the usual sum $f + f'$ and scalar multiplication $uf$ ($u \in \bbC$), the set of modular forms of weight $k$ with respect to $\Gamma$ forms a $\bbC$-vector space, denoted
\[
\mathcal{M}_k(\Gamma).
\]
The subset of cusp forms is a $\bbC$-vector subspace of $\mathcal{M}_k(\Gamma)$, denoted
\[
\calS_k(\Gamma).
\]
Moreover, the usual pointwise product of functions induces a graded-ring structure on
\[
\mathcal{M}(\Gamma) \defeq \bigoplus_{k \in \bbZ}\mathcal{M}_k(\Gamma),
\]
and
\[
\calS(\Gamma) \defeq \bigoplus_{k \in \bbZ}\calS_k(\Gamma)
\]
defines an ideal of $\mathcal{M}(\Gamma)$.
\end{example}
\subsection{Eisenstein Series}\label{Sec:MF_Eisenstein}
\begin{definition}
Let $k > 2$ be an even integer. The \termino{Eisenstein series of weight $k$} is the function
\[
G_k : \calH \to \bbC, \quad z \mapsto \sum_{(c, d) \in \bbZ^2 \setminus \{(0, 0)\}} \frac{1}{(cz + d)^k}.
\]
We write $\sideset{}{'}\sum_{(c, d)}$ for a sum over all $(c, d) \in \bbZ^2 \setminus \{(0, 0)\}$.
\end{definition}
\begin{proposition}\label{Prop:MF_Eisenstein}
Let $k > 2$ be an even integer. Then $G_k \in \mathcal{M}_k(\SL_2(\bbZ))$.
\end{proposition}
\begin{proof}
\leavevmode
\begin{enumerate}[label=(\arabic*)]
\item $G_k$ is holomorphic on $\calH$.
We prove absolute convergence on $\calH$ and uniform convergence on each region
\[
\Omega(A, B) \defeq \{z = x + iy \in \calH \mid |x| \leqslant A,\ y \geqslant B\}, \quad A, B > 0.
\]
We seek a constant $K > 0$, depending only on $A$ and $B$, such that
\begin{equation}\label{eq:MF_Eisenstein_bound}
|cz + d|^2 \geqslant K|ci + d|^2 = K(c^2 + d^2)
\end{equation}
for every $z \in \Omega(A, B)$ and $(c, d) \neq (0, 0)$. Take
\[
K = \frac{B^2}{1 + (A + B)^2} < 1.
\]
If $c = 0$, \eqref{eq:MF_Eisenstein_bound} follows from $K < 1$. If $c \neq 0$, put $\lambda = d/c$. It suffices to show that
\[
\frac{(\lambda + x)^2 + y^2}{1 + \lambda^2} \geqslant K.
\]
\begin{enumerate}[label=\textbf{Case~\arabic*}]
\item $|\lambda| \leqslant A + B$.
Since $y \geqslant B$,
\[
\frac{(\lambda + x)^2 + y^2}{1 + \lambda^2} \geqslant \frac{B^2}{1 + (A + B)^2} = K.
\]
\item $|\lambda| > A + B$.
We have
\[
\left|\frac{x}{\lambda}\right| \leqslant \frac{A}{|\lambda|} < \frac{A}{A + B} < 1.
\]
By the reverse triangle inequality,
\[
\left|1 + \frac{x}{\lambda}\right| \geqslant 1 - \left|\frac{x}{\lambda}\right| > \frac{B}{A + B}.
\]
% The board's next intermediate line omits an absolute value on lambda; correction pending confirmation. The squared estimate below is stated on the board.
Hence
\[
\frac{(\lambda + x)^2 + y^2}{1 + \lambda^2} \geqslant \frac{B^2}{(A + B)^2}\frac{\lambda^2}{1 + \lambda^2}.
\]
The function $t \mapsto t/(1 + t)$ is increasing for $t \geqslant 0$, so
\[
\frac{\lambda^2}{1 + \lambda^2} \geqslant \frac{(A + B)^2}{1 + (A + B)^2}.
\]
Combining these inequalities gives the required lower bound $K$.
\end{enumerate}
By \eqref{eq:MF_Eisenstein_bound},
\[
\frac{1}{|cz + d|^k} \leqslant \frac{K^{-k/2}}{|ci + d|^k}.
\]
The series
\[
\sideset{}{'}\sum_{(c, d)} \frac{1}{|ci + d|^k} = \sideset{}{'}\sum_{(c, d)} \frac{1}{(c^2 + d^2)^{k/2}}
\]
converges for $k > 2$; this is left as an exercise. The Weierstrass M-test gives absolute and uniform convergence on $\Omega(A, B)$. Every compact subset of $\calH$ is contained in some $\Omega(A, B)$, and each summand is holomorphic by \lemref{Lem:MF_denominator}. Therefore $G_k$ is holomorphic on $\calH$.
\item $G_k$ is weakly modular of weight $k$ with respect to $\SL_2(\bbZ)$.
Let $g = \begin{pmatrix}\alpha & \beta\\ \gamma & \delta\end{pmatrix} \in \SL_2(\bbZ)$. Then
\[
\begin{aligned}
G_k(gz) &= \sideset{}{'}\sum_{(c, d)} \frac{1}{\left(c\frac{\alpha z + \beta}{\gamma z + \delta} + d\right)^k}\\
&= (\gamma z + \delta)^k\sideset{}{'}\sum_{(c, d)} \frac{1}{((c\alpha + d\gamma)z + (c\beta + d\delta))^k}.
\end{aligned}
\]
The map
\[
\bbZ^2 \setminus \{(0, 0)\} \to \bbZ^2 \setminus \{(0, 0)\}, \quad (c, d) \mapsto (c, d)g = (c\alpha + d\gamma, c\beta + d\delta)
\]
is a bijection with inverse $(c, d) \mapsto (c, d)g^{-1}$. By absolute convergence, we may reindex the sum, obtaining
\[
G_k(gz) = (\gamma z + \delta)^kG_k(z),
\]
i.e. $G_k[g]_k = G_k$.
\item $G_k$ is holomorphic at $\infty$.
By (2), $G_k(z + 1) = G_k(z)$. Hence $G_k$ induces a holomorphic function $\widetilde G_k$ on the punctured unit disk through $q = e^{2\pi iz}$. It suffices to show that $\widetilde G_k$ is bounded near $q = 0$.
Fix $B > 0$. By (1), $G_k$ is bounded on $\Omega(1, B)$ by the finite constant
\[
K^{-k/2}\sideset{}{'}\sum_{(c, d)} \frac{1}{|ci + d|^k}.
\]
By periodicity, every $z$ with $\Im(z) \geqslant B$ can be translated into $\Omega(1, B)$ without changing $G_k(z)$. Thus $\widetilde G_k$ is bounded for $0 < |q| \leqslant e^{-2\pi B}$ and extends holomorphically across $q = 0$ by the removable singularity theorem. Finally, $G_k[\alpha]_k = G_k$ for every $\alpha \in \SL_2(\bbZ)$, so all the conditions in \defref{Def:MF_modular_form} hold.
\end{enumerate}
\end{proof}
\begin{proposition}[Fourier Expansion]\label{Prop:MF_Eisenstein_Fourier}
Let $k > 2$ be an even integer and $q = e^{2\pi iz}$. Then
\[
G_k(z) = 2\zeta(k) + \frac{2(2\pi i)^k}{(k - 1)!}\sum_{n = 1}^{\infty}\sigma_{k - 1}(n)q^n,
\]
where
\[
\zeta(k) = \sum_{m = 1}^{\infty}\frac{1}{m^k}, \qquad \sigma_{k - 1}(n) = \sum_{\substack{m \mid n\\m > 0}}m^{k - 1}.
\]
\end{proposition}
\begin{proof}
\leavevmode
\begin{enumerate}[label=\textit{Step~\arabic*.}]
\item We use the identities
\[
\pi\cot(\pi z) = \pi\frac{\cos(\pi z)}{\sin(\pi z)} = \frac{1}{z} + \sum_{m = 1}^{\infty}\left(\frac{1}{z + m} + \frac{1}{z - m}\right).
\]
The partial fraction expansion is quoted here; its justification is deferred to a later lecture. For $z \in \calH$, $|q| < 1$, and the exponential formulas for sine and cosine give
\[
\begin{aligned}
\pi\cot(\pi z) &= i\pi\frac{q + 1}{q - 1} = -i\pi\frac{1 + q}{1 - q}\\
&= -i\pi(1 + q)\sum_{n = 0}^{\infty}q^n = i\pi - 2\pi i\sum_{n = 0}^{\infty}q^n.
\end{aligned}
\]
Differentiating $k - 1$ times yields
\begin{equation}\label{eq:MF_cotangent_derivative}
\sum_{m \in \bbZ}\frac{1}{(m + z)^k} = \frac{(-2\pi i)^k}{(k - 1)!}\sum_{n = 1}^{\infty}n^{k - 1}q^n.
\end{equation}
\item Separate the terms with $c = 0$ and $c \neq 0$. Since $k$ is even,
\[
\begin{aligned}
G_k(z) &= \sum_{d \neq 0}\frac{1}{d^k} + \sum_{c \neq 0}\sum_{d \in \bbZ}\frac{1}{(cz + d)^k}\\
&= 2\zeta(k) + 2\sum_{c = 1}^{\infty}\sum_{d \in \bbZ}\frac{1}{(cz + d)^k}.
\end{aligned}
\]
By \eqref{eq:MF_cotangent_derivative} applied to $cz$,
\[
2\sum_{c = 1}^{\infty}\sum_{d \in \bbZ}\frac{1}{(cz + d)^k} = \frac{2(2\pi i)^k}{(k - 1)!}\sum_{c = 1}^{\infty}\sum_{m = 1}^{\infty}m^{k - 1}q^{cm}.
\]
Grouping equal powers of $q$ gives
\[
\sum_{c = 1}^{\infty}\sum_{m = 1}^{\infty}m^{k - 1}q^{cm} = \sum_{n = 1}^{\infty}\left(\sum_{\substack{m \mid n\\m > 0}}m^{k - 1}\right)q^n = \sum_{n = 1}^{\infty}\sigma_{k - 1}(n)q^n.
\]
This proves the formula.
\end{enumerate}
\end{proof}
\subsection{The Discriminant and the Modular Invariant}\label{Sec:MF_discriminant}
\begin{definition}
Define
\[
g_2(z) \defeq 60G_4(z), \qquad g_3(z) \defeq 140G_6(z).
\]
The \termino{discriminant function} is
\[
\Delta : \calH \to \bbC, \quad z \mapsto g_2(z)^3 - 27g_3(z)^2.
\]
\end{definition}
Since $G_4 \in \mathcal{M}_4(\SL_2(\bbZ))$, $G_6 \in \mathcal{M}_6(\SL_2(\bbZ))$ and $\mathcal{M}(\SL_2(\bbZ))$ is a graded ring, we have
\[
\Delta \in \mathcal{M}_{12}(\SL_2(\bbZ)).
\]
\begin{remarkstar}
Using the expression for $\zeta(2k)$ in terms of Bernoulli numbers, verify that $\Delta(\infty) = 0$. Thus $\Delta \in \calS_{12}(\SL_2(\bbZ))$. This is left as an exercise.
\end{remarkstar}
\begin{remark}
We will later prove that $\Delta(z) \neq 0$ for every $z \in \calH$. In particular, $\Delta$ is a nonzero cusp form of weight $12$.
\end{remark}
\begin{definition}
The \termino{modular invariant} is the function
\[
j : \calH \to \bbC, \quad z \mapsto 1728\frac{g_2(z)^3}{\Delta(z)}.
\]
\end{definition}
\begin{remark}
Using the nonvanishing of $\Delta$ on $\calH$, the properties of $G_4$ and $\Delta$ show that $j$ is holomorphic on $\calH$ and invariant under $\SL_2(\bbZ)$. We will also see later that $j$ has a simple pole at $\infty$.
\end{remark}
\begin{remarkstar}
A goal of the course is to compute $\dim\mathcal{M}_k(\Gamma)$ and relate modular forms to the geometry of the modular curve $X(\Gamma)$, complex tori and elliptic curves.
\end{remarkstar}