forked from patmorin/turan
-
Notifications
You must be signed in to change notification settings - Fork 0
/
slides.tex
465 lines (391 loc) · 15 KB
/
slides.tex
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
\documentclass{beamer}
\usepackage{amsmath}
\usepackage{tipa}
\usepackage[utf8]{inputenc}
\usepackage{xmpmulti}
\mode<presentation>{
\definecolor{cured}{rgb}{.8,0,.2}
\usecolortheme[named=cured]{structure}
\usetheme{split}
}
\newcommand{\taco}{\raisebox{-.1ex}{\includegraphics[height=2.0ex]{figs/triangles-edge-1}}}
\newcommand{\mariposa}{\raisebox{-.1ex}{\includegraphics[height=2.0ex]{figs/triangles-edge-2}}}
\newcommand{\bat}{\raisebox{-.1ex}{\includegraphics[height=2.0ex]{figs/triangles-vertex-1}}}
\newcommand{\nested}{\raisebox{-.1ex}{\includegraphics[height=2.0ex]{figs/triangles-vertex-2}}}
\newcommand{\crossing}{\raisebox{-.1ex}{\includegraphics[height=2.0ex]{figs/triangles-vertex-3}}}
\newcommand{\ears}{\raisebox{-.1ex}{\includegraphics[height=2.0ex]{figs/triangles-disjoint-1}}}
\newcommand{\swords}{\raisebox{-.1ex}{\includegraphics[height=2.0ex]{figs/triangles-disjoint-2}}}
\newcommand{\david}{\raisebox{-.1ex}{\includegraphics[height=2.0ex]{figs/triangles-disjoint-3}}}
\newcommand{\mi}[1]{\multiinclude[<+>][start=1,format=pdf]{#1}}
\DeclareMathOperator{\ex}{ex}
\input{pat-slides}
\title{Turán-Type Theorems for Triangles \newline in Convex Point Sets}
\author{Pat Morin}
\date{Shonan Village Meeting \\ May 30, 2016}
\begin{document}
\begin{frame}
\titlepage
\end{frame}
\begin{frame}
\frametitle{$\ex(n,\{\taco,\nested\})$}
% \framesubtitle{
% %\newlength{\ka}
% \setlength{\ka}{.3\textwidth}
% \addtolength{\ka}{-1cm}
% \begin{tabular}{c@{\hspace{1cm}}c}
% \includegraphics[width=.48\ka]{figs/crapper-2} &
% \includegraphics[width=.48\ka]{figs/crapper-1}
% \end{tabular}
% }
\begin{itemize}
%\item<1-> We know $\ex'(n,X)$ for every $X$ of size 2 except
% $X=\{\taco,\nested\}$
%\item<1-> Posed in Bra\ss\ 2004
\item<1-> A puzzle played in $n$ rounds on the $n\times n$ grid
\item<2-> Within each round, points must form an increasing sequence
\item<3-> Each point kills a {\raisebox{-.1ex}{\includegraphics[angle=180,origin=c,height=2.0ex]{slidefigs/shape}}}
shape that can't be played in subsequent rounds
\item<4-> The goal is to play as many points as possible
% \item<4-> $\Omega(n^{3/2})$ lower bound is easy (and looks tight):\\
% \begin{center}
% \only<1-4>{\includegraphics{slidefigs/n32-1}}%
% \only<5>{\includegraphics{slidefigs/n32-2}}%
% \only<6>{\includegraphics{slidefigs/n32-3}}%
% \only<7>{\includegraphics{slidefigs/n32-4}}%
% \only<8>{\includegraphics{slidefigs/n32-5}}%
% \only<9>{\includegraphics{slidefigs/n32-6}}%
% \only<10->{\includegraphics{slidefigs/n32-7}}%
% \end{center}
%\item<11-> We have no upper bound better than $O(n^2)$
\end{itemize}
\end{frame}
\begin{frame}
\frametitle{$\ex(n,\{\taco,\nested\})\in\omega(n^{3/2})$}
\begin{itemize}
%\item Ignoring constant factors, we can assume a $n\times n$ board
%\item Rotating, each point kills an L shape
\centerline{\mi{slidefigs/sqrtn}}%
\item $\only<2->{3}
\only<4->{+3}
\only<6->{+3}
\only<8->{+3}
\only<10->{+3}
\only<12->{+3}
\only<14->{+3}
\only<16->{+3}
\only<18->{+3}
\only<19->{=27=9^{3/2}}
$
\end{itemize}
\end{frame}
\begin{frame}
\frametitle{$\ex(n,\{\taco,\nested\})\in\omega(n^{3/2})$}
\begin{itemize}
\item Another strategy:
\centerline{\mi{slidefigs/sqrtn-david}}%
\item $\only<2->{3}
\only<3->{+3}
\only<4->{+3}
\only<5->{+3}
\only<6->{+3}
\only<7->{+3}
\only<8->{+3}
\only<9->{+3}
\only<10->{+3}
\only<11->{=27=9^{3/2}}
$
\end{itemize}
\end{frame}
\begin{frame}
\frametitle{$\ex(n,\{\taco,\nested\})\in\omega(n^{3/2})$}
\begin{itemize}
\item A third try:
\centerline{\mi{slidefigs/nine-twentyeight}}%
\item $\only<2->{3}
\only<3->{+3}
\only<4->{+3}
\only<5->{+3}
\only<6->{+{\color{red}4}}
\only<7->{+3}
\only<8->{+3}
\only<9->{+3}
\only<10->{+3}
\only<11->{=28=9^{1.516551628}}
$
\end{itemize}
\end{frame}
\begin{frame}
\frametitle{$\ex(n,\{\taco,\nested\})\in\omega(n^{3/2})$}
\begin{itemize}
\item For general $n$, use recursion on subproblems of size $(n/9)\times(n/9)$
\centerline{\includegraphics[scale=0.8]{slidefigs/recursion}}%
\item[] \[ T(n) = \begin{cases}
28 & \text{if $n=9$} \\
28 T(n/9) & \text{if $n=9^k$} \end{cases} \]
\item $ T(n) = \Theta(n^{\log_9 28}) = \Theta(n^{1.516551628})$
\end{itemize}
\end{frame}
\begin{frame}
\frametitle{$\ex(n,\{\taco,\nested\})$}
\begin{itemize}[<+->]
\item Can we do better than $\Theta(n^{1.516551628})$?
\item $70\times70$ game has an $652$ point solution
\item Recursion gives $\Omega(n^{1.5252564009\ldots})$ solution for $n\times n$
\item That's the best lower-bound we have so far\ldots
\end{itemize}
\end{frame}
\begin{frame}
\frametitle{$\ex(n,\{\taco,\nested\})$---Upper Bounds}
\begin{itemize}[<+->]
\item $O(n^2)$ upper bound is trivial.
\item Treat $x$ and $y$ coordinates as vertices of a bipartite graph, $G$
\item Each round, $i$, becomes a matching, $M_i$
\item No edge from $M_j$, $j\neq i$, joins two vertices used in $M_i$. \\[1ex]
\centerline{\includegraphics{slidefigs/induced}}
\item Puzzle solution gives a solution to the \emph{Rusza-Szemerédi Induced Matching Problem}
\begin{itemize}
\item Puzzle score $=|E(G)|$
\end{itemize}
\item IM Problem has an $O(n^{2/e^{\log^* n}})$ upper bound
\item Maximum puzzle score is $O(n^2/e^{\log^* n})$
\end{itemize}
\end{frame}
\begin{frame}
\frametitle{Maximum Triangles Problem}
\framesubtitle{Erd\H{o}s and Purdy (1971)}
\begin{itemize}
\item What is the largest number of maximum area triangles determined by a set of $n$ points?\\[1ex]
\centerline{\mi{slidefigs/erdos-regular}}%
\item<2->Answer: At least $n$
\end{itemize}
\end{frame}
\begin{frame}
\frametitle{Maximum Triangles Problem}
\framesubtitle{Bra\ss, Rote, Swanepoel (2001)}
\only<+->{
\textbf{Theorem:} Any set of $n$ points determines at most $n$ maximum-area triangles}
\begin{itemize}[<+->]
\item WLOG we can assume points are in convex position
\item How can a pair of maximum area triangles interleave?
\begin{itemize}
\item Boyce, Dobkin, Drysdale, Guibas (1985)
\item Like this: \taco, \david, \crossing, \mariposa
\item Not like this: \ears, \swords, \nested, \bat
\end{itemize}
\item What is the maximum number of triangles we can draw on a convex
$n$-gon before we
get a pair like \ears, \swords, \nested, or \bat?
\item No geometry!
\end{itemize}
\end{frame}
\begin{frame}
\frametitle{Maximum Triangles Problem}
\framesubtitle{Bra\ss, Rote, Swanepoel (2001)}
Let $\ex(n,X)$ denote the maximum number of triangles we can draw on a convex $n$-gon before we get a configuration from $X$.\\[2ex]
\uncover<2->
{\textbf{Theorem (BRS2001):} $\ex(n,\{\ears,\swords,\nested,\bat\})\le n$.}
% \begin{itemize}
% \item<3-> Number polygon vertices $0,\ldots,n-1$
% \item<4-> Represent each triangle as a triple: $(a,b,c)$, with $0\le a<b<c<n$.
% \item<5-> Partially-order triangles according to
% \[ (x_1,y_1,z_1)\preceq (x_2,y_2,z_2) \Leftrightarrow x_1\le x_2, y_1\le y_2, z_1\le z_2 \]
% \item<6-> Key insight: Any pair of maximum-area triangles (\taco, \david, \crossing, \mariposa) is comparable!
% \item<7-> $\Delta_1<\Delta_2<\cdots<\Delta_k \Rightarrow k < 3n$
% \item<8-> Count a bit more carefully to get $k\le n$ \hfill{QED}
% \end{itemize}
\end{frame}
\begin{frame}
\frametitle{Turán-Type Theorems for Triangles in Convex Point Sets}
\framesubtitle{256 Problems}
\begin{itemize}[<+->]
\item $\ex(n,X)$ is the maximum number of triangles we can draw on
a convex $n$-gon before we get a configuration from $X$
\item 256 problems: Determine $\ex(n,X)$ for every $X\subseteq\{\taco,\david,\crossing,\mariposa,\ears,\swords,\nested,\bat\}$
\end{itemize}
\end{frame}
\begin{frame}
\frametitle{Bra\ss\ 2004}
Bra\ss\ (2004):
\[ \begin{array}{rrr}
\ex(n,\{\ears\})=\Theta(n^3) & \ex(n,\{\swords\})=\Theta(n^2) & \ex(n,\{\david\})=\Theta(n^2) \\
\ex(n,\{\bat\})=\Theta(n^3) & \ex(n,\{\nested\})=\Theta(n^2) & \ex(n,\{\crossing\})=\Theta(n^2) \\
\ex(n,\{\mariposa\})=\Theta(n^3) & \ex(n,\{\taco\})=\Theta(n^2)
\end{array}
\]
\[\begin{array}{r}
\ex(n,\{\bat,\nested\})=\Theta(n^2) \\ \ex(n,\{\crossing,\nested\})=\Theta(n^2) \\ \ex(n,\{\bat,\crossing\})=\Theta(n^2) \\
\end{array}\]
\end{frame}
\begin{frame}
\frametitle{Proof that $\ex(n,\{\crossing\})=O(n^2)$}
\framesubtitle{Bra\ss\ (2004)}
\centerline{\mi{slidefigs/crossing}}
\uncover<4->{\hfill{QED}}
\end{frame}
\begin{frame}
\frametitle{Proof that $\ex(n,\{\swords\})=O(n^2)$}
\framesubtitle{Bra\ss (2004)}
\centerline{\mi{slidefigs/swords}}
\begin{itemize}[<4->]
\item[]\ \hfill{QED}
\end{itemize}
\end{frame}
\begin{frame}
\frametitle{Proof that $\mariposa$ doesn't matter}
\textbf{Theorem:} Any set of triangles of size $N$ contains a subset
of size at least $N/8$ that has no \mariposa\ pair
\begin{itemize}
\item<2-> Randomly direct each pair $(u,v)$ of vertices ($\overrightarrow{uv}$ or $\overleftarrow{uv}$)
\item<3-> Discard all triangles that use $uv$ and are on the right of $uv$
\end{itemize}
\centerline{
\only<1>{\includegraphics{slidefigs/chucker-1}}%
\only<2>{\includegraphics{slidefigs/chucker-2}}%
\only<3->{\includegraphics{slidefigs/chucker-3}}%
}
\begin{itemize}[<4->]
\item Resulting set has expected size $N/8$ and no $\mariposa$ pairs
\hfill{QED}
\end{itemize}
\end{frame}
\begin{frame}
\frametitle{The Top-Bottom View}
Define $\ex'(n,X)$ just like $\ex(n,X)$ but only count triangles
with one vertex in the bottom half and two in the top half.
\centerline{\includegraphics{slidefigs/top-bottom}}
\uncover<2->{
\textbf{Lemma:}
If $\ex'(n,X)\in O(n^c)$, then
\[
\ex(n,X)\in
\begin{cases}
O(n^c) & \text{if $c>1$} \\
O(n\log n) & \text{if $c=1$}
\end{cases}
\]
}
\uncover<3>{
\emph{Proof:} $\ex(n)\le 2\ex'(n) + 2\ex(n/2)$
}
\end{frame}
\begin{frame}
\frametitle{The Dot-Puzzle View}
\framesubtitle{A One-Player Game}
\includegraphics[width=.95\textwidth]{figs/point-view}
\uncover<2->{\begin{center}A puzzle played in $n$ rounds\end{center}}
\end{frame}
\begin{frame}
\frametitle{The Dot-Puzzle View}
\framesubtitle{The Rules}
\begin{center}
\newlength{\ka}
\setlength{\ka}{\textwidth}
\addtolength{\ka}{-1cm}
\begin{tabular}{c@{\hspace{1cm}}c}
\includegraphics[width=.48\ka]{figs/crapper-2} &
\includegraphics[width=.48\ka]{figs/crapper-1} \\
within round & subsequent rounds
\end{tabular}
\end{center}
\end{frame}
\begin{frame}
\frametitle{\swords\ is a killer}
\framesubtitle{
%\newlength{\ka}
\setlength{\ka}{.3\textwidth}
\addtolength{\ka}{-1cm}
\begin{tabular}{c@{\hspace{1cm}}c}
\includegraphics[width=.48\ka]{figs/crapper-2} &
\includegraphics[width=.48\ka]{figs/crapper-1}
\end{tabular}
}
\centerline{\includegraphics{slidefigs/swords-region-1}}
\invisible<1->{\hfill{QED}}
\end{frame}
\begin{frame}
\frametitle{$\ex'(n,\{\swords,\taco\})=O(n)$}
\framesubtitle{
%\newlength{\ka}
\setlength{\ka}{.3\textwidth}
\addtolength{\ka}{-1cm}
\begin{tabular}{c@{\hspace{1cm}}c}
\includegraphics[width=.48\ka]{figs/crapper-2} &
\includegraphics[width=.48\ka]{figs/crapper-1}
\end{tabular}
}
\centerline{
\only<1>{\includegraphics{slidefigs/swords-region-1}}%
\only<2>{\includegraphics{slidefigs/swords-region-2}}%
\only<3>{\includegraphics{slidefigs/swords-region-3}}%
\only<4->{\includegraphics{slidefigs/swords-region-4}}%
}
\uncover<4->{\hfill{QED}}\\
\uncover<5->{$\ex'(n,\{\swords,\nested\})=O(n)$ and
$\ex'(n,\{\swords,\crossing\})=O(n)$ have similar proofs.}
\end{frame}
\begin{frame}
\frametitle{$\ex'(n,\{\taco,\nested,\crossing\}) = O(n)$}
\framesubtitle{
%\newlength{\ka}
\setlength{\ka}{.3\textwidth}
\addtolength{\ka}{-1cm}
\begin{tabular}{c@{\hspace{1cm}}c}
\includegraphics[width=.48\ka]{figs/crapper-2} &
\includegraphics[width=.48\ka]{figs/crapper-1}
\end{tabular}
}
\begin{itemize}[<+->]
\item \nested\ ensures points in one round are non-decreasing
\begin{itemize}
\item Each point is in a new row or new column
\end{itemize}
\item Union of rules ensures these points are killed in subsequent rounds:\\[1ex]
\centerline{\includegraphics{figs/killers-10}}
\item Each point in round $i$ kills one row or column in subsequent rounds\hfill{QED}
\end{itemize}
\end{frame}
\begin{frame}
\frametitle{$\ex'(n,\{\taco,\nested\}) = O(n)$}
\framesubtitle{
%\newlength{\ka}
\setlength{\ka}{.3\textwidth}
\addtolength{\ka}{-1cm}
\begin{tabular}{c@{\hspace{1cm}}c}
\includegraphics[width=.48\ka]{figs/crapper-2} &
\includegraphics[width=.48\ka]{figs/crapper-1}
\end{tabular}
}
\begin{itemize}[<+->]
\item Using these tools we have determined $\ex'(n,X)$ for any
two element set $X$ \emph{except} $X=\{\taco,\nested\}$.
\item The rules for $\ex'(n,X)$ for $\ex'(n,\{\taco,\nested\}$ give our
dot-puzzle game.
\item Did Bra\ss\ know this already in 2004?
\end{itemize}
\end{frame}
\begin{frame}
\frametitle{The End}
\begin{center}
%\newlength{\ka}
\setlength{\ka}{.6\textwidth}
\addtolength{\ka}{-1cm}
\begin{tabular}{c@{\hspace{1cm}}c}
\includegraphics[width=.48\ka]{figs/crapper-2} &
\includegraphics[width=.48\ka]{figs/crapper-1}
\end{tabular}
\end{center}
\centerline{\Huge Thanks!}
\end{frame}
\begin{frame}
\frametitle{References}
\begin{itemize}
\item P. Braß. Tur\'an-type extremal problems for convex geometric
hypergraphs. Contemporary Mathematics, \textbf{342}:25–34, 2004.
\item P. Braß, G. Rote, and K. J. Swanepoel. Triangles of extremal area
or perimeter in a finite planar point set. Discrete \& Computational
Geometry, \textbf{26}(1):51–58, 2001.
\item P. Erd\H{o}s and G. Purdy, Some extremal problems in geometry,
Journal of Combinatorial Theory Series A, \textbf{10}:246–252, 1971.
\end{itemize}
\end{frame}
\end{document}