-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathindex.html
More file actions
478 lines (412 loc) · 23.2 KB
/
index.html
File metadata and controls
478 lines (412 loc) · 23.2 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
<style>
.blur {
filter: blur(9px);
}
</style>
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1, shrink-to-fit=no">
<meta name="description" content="">
<meta name="author" content="">
<link href="https://fonts.googleapis.com/css?family=Raleway:100,200,300,400,500,600,700,800,900&display=swap" rel="stylesheet">
<title>MiNT</title>
<!--
Breezed Template
https://templatemo.com/tm-543-breezed
-->
<!-- Additional CSS Files -->
<link rel="stylesheet" type="text/css" href="assets/css/bootstrap.min.css">
<link rel="stylesheet" type="text/css" href="assets/css/font-awesome.css">
<link rel="stylesheet" href="assets/css/templatemo-breezed.css">
<link rel="stylesheet" href="assets/css/owl-carousel.css">
<link rel="stylesheet" href="assets/css/lightbox.css">
<link rel="stylesheet" href="assets/css/style.css">
</head>
<body>
<!-- ***** Preloader Start ***** -->
<div id="preloader">
<div class="jumper">
<div></div>
<div></div>
<div></div>
</div>
</div>
<!-- ***** Preloader End ***** -->
<!-- ***** Header Area Start ***** -->
<header class="header-area header-sticky">
<div class="container">
<div class="row">
<div class="col-12">
<nav class="main-nav">
<!-- ***** Logo Start ***** -->
<a href="index.html" class="logo">
MiNT
</a>
<!-- ***** Logo End ***** -->
<!-- ***** Menu Start ***** -->
<ul class="nav">
<li class="scroll-to-section"><a href="#top" class="active">Home</a></li>
<li class="scroll-to-section"><a href="#benchmark_impl">Benchmark</a></li>
<li class="scroll-to-section"><a href="#data_extraction">Data</a></li>
<li class="scroll-to-section"><a href="#metadata">Metadata</a></li>
<!-- <li class="submenu">-->
<!-- <a href="javascript:;">Drop Down</a>-->
<!-- <ul>-->
<!-- <li><a href="">About Us</a></li>-->
<!-- <li><a href="">Features</a></li>-->
<!-- <li><a href="">FAQ's</a></li>-->
<!-- <li><a href="">Blog</a></li>-->
<!-- </ul>-->
<!-- </li>-->
<!-- <li class="scroll-to-section"><a href="#contact-us">Contact Us</a></li> -->
<!-- <div class="search-icon">-->
<!-- <a href="#search"><i class="fa fa-search"></i></a>-->
<!-- </div>-->
</ul>
<a class='menu-trigger'>
<span>Menu</span>
</a>
<!-- ***** Menu End ***** -->
</nav>
</div>
</div>
</div>
</header>
<!-- ***** Header Area End ***** -->
<!-- ***** Search Area ***** -->
<div id="search">
<button type="button" class="close">×</button>
<form id="contact" action="#" method="get">
<fieldset>
<input type="search" name="q" placeholder="SEARCH KEYWORD(s)" aria-label="Search through site content">
</fieldset>
<fieldset>
<button type="submit" class="main-button">Search</button>
</fieldset>
</form>
</div>
<!-- ***** Main Banner Area Start ***** -->
<div class="main-banner header-text" id="top">
<div class="Modern-Slider">
<!-- Item -->
<div class="item">
<div class= " img-fill main-background", height>
<!-- img-fill -->
<!-- <img src="assets/images/n_light.gif" alt="" class="blur"> -->
<img src="assets/images/DTDG_background.png" alt="" class="blur">
<div class="text-content">
<h3>MiNT: Multi-Network Transfer Benchmark for Temporal Graph Learning</h3>
<h5>Multi-network training Benchmark</h5>
<a href="https://zenodo.org/records/15364297" class="main-filled-button">Dataset</a>
<a href="https://github.com/benjaminnNgo/ScalingTGNs" class="main-stroked-button">Benchmark Implementation</a>
<!-- <a href="https://huggingface.co/api/datasets/ntgbaoo/Temporal_Graph_Scaling_TGS_Benchmark/croissant" class="main-filled-button">Dataset ML Croissant</a> -->
</div>
</div>
</div>
<!-- // Item -->
<!-- Item -->
<!-- <div class="item">-->
<!-- <div class="img-fill">-->
<!-- <img src="assets/images/slide-02.jpg" alt="">-->
<!-- <div class="text-content">-->
<!-- <h3>Integrated Marketing Media</h3>-->
<!-- <h5>Best Digital Marketing</h5>-->
<!-- <a href="#" class="main-stroked-button">Read More</a>-->
<!-- <a href="#" class="main-filled-button">Take Action</a>-->
<!-- </div>-->
<!-- </div>-->
<!-- </div>-->
<!-- <!– // Item –>-->
<!-- <!– Item –>-->
<!-- <div class="item">-->
<!-- <div class="img-fill">-->
<!-- <img src="assets/images/slide-03.jpg" alt="">-->
<!-- <div class="text-content">-->
<!-- <h3>High Performance</h3>-->
<!-- <h5>Robust and Speedy</h5>-->
<!-- <a href="#" class="main-stroked-button">Learn More</a>-->
<!-- <a href="#" class="main-filled-button">Get It Now</a>-->
<!-- </div>-->
<!-- </div>-->
<!-- </div>-->
<!-- // Item -->
</div>
</div>
<!-- <div class="scroll-down scroll-to-section"><a href="#about"><i class="fa fa-arrow-down"></i></a></div> -->
<!-- ***** Main Banner Area End ***** -->
<!-- ***** Introduction Area Starts ***** -->
<section class="section" id="introduction">
<div class="container">
<div class="row">
<div class="col-lg-6 col-md-6 col-xs-12">
<div class="left-text-content">
<div class="section-heading">
<h6>Introduction</h6>
<h2></h2>
</div>
<p>MiNT provides the implementation of the first Multi-network training benchmarking. We introduce MiNT-train, the first algorithm to train TGNNs across multiple networks simultaneously, expanding possibilities for temporal graph learning.</p>
<!-- <img src="assets/images/TGS_Logo.png" alt="" style="max-width: 60%;
height: auto; alignment: center; text-align: center; vertical-align: middle;"> -->
</div>
</div>
<div class="col-lg-6 col-md-6 col-xs-12">
<div class="right-text-content">
<p>
Temporal Graph Learning (TGL) aims to discover patterns on evolving networks or temporal graphs, and leverage such patterns to predict future interactions. However, most existing work focuses on learning from single networks in isolation, leaving the question of cross-network generalization largely unexplored. In this study, we introduce a new benchmark of 84 real-world temporal transaction networks and propose Temporal Multi-network Training (MiNT), a pre-training framework designed to capture transferable temporal dynamics across diverse networks. We train MiNT models on up to 64 transaction networks and evaluate their generalization ability on 20 held-out, unseen networks. Our results show that MiNT consistently outperforms individually trained models, revealing a strong relation between the number of pre-training networks and transfer performance. These findings highlight scaling trends in temporal graph learning and underscore the importance of network diversity in improving generalization. This work establishes the first large-scale benchmark for studying transferability in TGL and lays the groundwork for developing Temporal Graph Foundation Models.
</div>
</div>
</div>
</div>
</section>
<!-- ***** Introduction Area Ends ***** -->
<!-- ***** Features Big Item Start ***** -->
<section class="section" id="benchmark_impl">
<div class="container">
<div class="row">
<h6>MiNT Benchmark</h6>
<p style="margin-top: 10px;">transaction networks are divided randomly into train and test sets. The train set is used to train foundation models with different sizes; then, the trained models are evaluated on the test set. Test AUC of foundation models trained on 4, 16 and 64 networks and evaluated on unseen test datasets. We compare the performance with persistence forecast, and single models trained and tested on each dataset. </p>
<img src="assets/images/Figure4_mint_wo_pf.jpg" alt="" style="max-width: 100%;
height: auto; alignment: center; text-align: center;">
</div>
</div>
</section>
<!-- ***** Features Big Item End ***** -->
<!-- ***** About Area Starts ***** -->
<section class="section" id="data_extraction">
<div class="container">
<div class="row">
<div class="col-lg-6 col-md-6 col-xs-12">
<div class="left-text-content">
<div class="section-heading">
<h6>MiNT System Overview</h6>
<h2>MiNT has three main steps</h2>
</div>
<p>
<ol>
<li>(1) Token extraction: extracting the token transaction network from our P2P Ethereum live node.</li>
<li>(2) Discretization: creating weekly snapshots to form discrete time dynamic graphs.</li>
<!-- <li>(3) Foundation Model Training: TGS transaction networks are divided randomly into train and test sets. We train the FMs on a collection of training networks. Lastly, FMs are tested on 20 unseen test networks.</li> -->
</ol>
</p>
</div>
</div>
<div class="col-lg-6 col-md-6 col-xs-12">
<div class="right-text-content">
<img src="assets/images/System_overview_v3.png" alt="" style="max-width: 100%;
height: auto; alignment: center; text-align: center; vertical-align: middle;">
</div>
</div>
</div>
</div>
</section>
<!-- ***** About Area Ends ***** -->
<!-- ***** Features Big Item Start ***** -->
<section class="section" id="metadata">
<div class="container">
<div class="row">
<h6>Metadata</h6>
<p>
Datasets provided in this storage are introduced in the paper: MiNT: Multi-Network Transfer Benchmark for Temporal Graph Learning
Each .csv file represents all transactions of the token network that has the same name as the file name (tokenname.csv)
Each transaction corresponds to a row in each file. <br><br></p>
<ol>
<li> blockNumber : is the block ID of Ethereum that includes this transaction</li>
<li>timestamp: time that the transaction is made in UNIX timestamp format</li>
<li>tokenAddress : the address that specifies a unique ERC20 token</li>
<li>from: address of sender</li>
<li>to: address of receiver</li>
<li> value: the amount the transaction</li>
<li>fileBlock: we split the whole number of blocks count to 35 buckets and assigned the bucket ID to the transaction to trace the blocks</li>
</ol>
<br>
<p>
<br><br>
Raw .csv will be used to divide into generate edgeslist and label, which indicates all node interactions and labels for each snapshot respectively, with help of functions from TGS_Handler defined in TGS.py inside the TGS package (see the code in the Github repository provided along with this data storage)
</p>
<pre>
<br>
Data Example:
blockNumber,timestamp,tokenAddress,from,to,value,fileBlock
16480746,1674612419,0x5a3e6a77ba2f983ec0d371ea3b475f8bc0811ad5,0x0000000000000000000000000000000000000000,0x0000ba9ff5c97f33bd62c216a56b3d02ae6ac4bb,1E+18,26
</pre>
</div>
</div>
</section>
<!-- ***** Features Big Item End ***** -->
<!-- <!– ***** Testimonials Starts ***** –>-->
<!-- <section class="section" id="testimonials">-->
<!-- <div class="container">-->
<!-- <div class="row">-->
<!-- <div class="col-lg-12">-->
<!-- <div class="section-heading">-->
<!-- <h6>Digital Team</h6>-->
<!-- <h2>young and talented members</h2>-->
<!-- </div>-->
<!-- </div>-->
<!-- <div class="col-lg-12 col-md-12 col-sm-12 mobile-bottom-fix-big" data-scroll-reveal="enter left move 30px over 0.6s after 0.4s">-->
<!-- <div class="owl-carousel owl-theme">-->
<!-- <div class="item author-item">-->
<!-- <div class="member-thumb">-->
<!-- <img src="assets/images/mcgill-university-logo-png-transparent-cropped.png" alt="">-->
<!-- <div class="hover-effect">-->
<!-- <div class="hover-content">-->
<!-- </div>-->
<!-- </div>-->
<!-- </div>-->
<!-- <h4>.01 White Cheese</h4>-->
<!-- <span>Digital Marketer</span>-->
<!-- </div>-->
<!-- <div class="item author-item">-->
<!-- <div class="member-thumb">-->
<!-- <img src="assets/images/Mila_Logo_Web_Coul_RGB.png" alt="">-->
<!-- <div class="hover-effect">-->
<!-- <div class="hover-content">-->
<!-- </div>-->
<!-- </div>-->
<!-- </div>-->
<!-- <h4>.02 Snow Mary</h4>-->
<!-- <span>Site Analyst</span>-->
<!-- </div>-->
<!-- <div class="item author-item">-->
<!-- <div class="member-thumb">-->
<!-- <img src="assets/images/ucf-university-of-central-florida-logo-503B602697-seeklogo.com.png" alt="">-->
<!-- <div class="hover-effect">-->
<!-- <div class="hover-content">-->
<!-- <ul>-->
<!-- <li><a href="#"><i class="fa fa-facebook"></i></a></li>-->
<!-- <li><a href="#"><i class="fa fa-twitter"></i></a></li>-->
<!-- <li><a href="#"><i class="fa fa-linkedin"></i></a></li>-->
<!-- <li><a href="#"><i class="fa fa-rss"></i></a></li>-->
<!-- </ul>-->
<!-- </div>-->
<!-- </div>-->
<!-- </div>-->
<!-- <h4>.03 Johnny Egg</h4>-->
<!-- <span>Digital Influencer</span>-->
<!-- </div>-->
<!-- <div class="item author-item">-->
<!-- <div class="member-thumb">-->
<!-- <img src="assets/images/member-item-05.jpg" alt="">-->
<!-- <div class="hover-effect">-->
<!-- <div class="hover-content">-->
<!-- <ul>-->
<!-- <li><a href="#"><i class="fa fa-facebook"></i></a></li>-->
<!-- <li><a href="#"><i class="fa fa-twitter"></i></a></li>-->
<!-- <li><a href="#"><i class="fa fa-linkedin"></i></a></li>-->
<!-- <li><a href="#"><i class="fa fa-rss"></i></a></li>-->
<!-- </ul>-->
<!-- </div>-->
<!-- </div>-->
<!-- </div>-->
<!-- <h4>.04 Catherine Phyu</h4>-->
<!-- <span>Co Founder</span>-->
<!-- </div>-->
<!-- <div class="item author-item">-->
<!-- <div class="member-thumb">-->
<!-- <img src="assets/images/member-item-03.jpg" alt="">-->
<!-- <div class="hover-effect">-->
<!-- <div class="hover-content">-->
<!-- <ul>-->
<!-- <li><a href="#"><i class="fa fa-facebook"></i></a></li>-->
<!-- <li><a href="#"><i class="fa fa-twitter"></i></a></li>-->
<!-- <li><a href="#"><i class="fa fa-linkedin"></i></a></li>-->
<!-- <li><a href="#"><i class="fa fa-rss"></i></a></li>-->
<!-- </ul>-->
<!-- </div>-->
<!-- </div>-->
<!-- </div>-->
<!-- <h4>.05 Shao Lynn</h4>-->
<!-- <span>Chief Marketing</span>-->
<!-- </div>-->
<!-- <div class="item author-item">-->
<!-- <div class="member-thumb">-->
<!-- <img src="assets/images/member-item-04.jpg" alt="">-->
<!-- <div class="hover-effect">-->
<!-- <div class="hover-content">-->
<!-- <ul>-->
<!-- <li><a href="#"><i class="fa fa-facebook"></i></a></li>-->
<!-- <li><a href="#"><i class="fa fa-twitter"></i></a></li>-->
<!-- <li><a href="#"><i class="fa fa-linkedin"></i></a></li>-->
<!-- <li><a href="#"><i class="fa fa-rss"></i></a></li>-->
<!-- </ul>-->
<!-- </div>-->
<!-- </div>-->
<!-- </div>-->
<!-- <h4>.06 Emma Honey</h4>-->
<!-- <span>Digital Influencer</span>-->
<!-- </div>-->
<!-- <div class="item author-item">-->
<!-- <div class="member-thumb">-->
<!-- <img src="assets/images/member-item-06.jpg" alt="">-->
<!-- <div class="hover-effect">-->
<!-- <div class="hover-content">-->
<!-- <ul>-->
<!-- <li><a href="#"><i class="fa fa-facebook"></i></a></li>-->
<!-- <li><a href="#"><i class="fa fa-twitter"></i></a></li>-->
<!-- <li><a href="#"><i class="fa fa-linkedin"></i></a></li>-->
<!-- <li><a href="#"><i class="fa fa-rss"></i></a></li>-->
<!-- </ul>-->
<!-- </div>-->
<!-- </div>-->
<!-- </div>-->
<!-- <h4>.07 Oliva Sofie</h4>-->
<!-- <span>Website Analyst</span>-->
<!-- </div>-->
<!-- -->
<!-- </div>-->
<!-- </div>-->
<!-- </div>-->
<!-- </div>-->
<!-- </section>-->
<!-- <!– ***** Testimonials Ends ***** –>-->
<!-- -->
<!-- ***** Footer Start ***** -->
<footer>
<div class="container">
<div class="row">
<div class="col-lg-6 col-xs-12">
<div class="left-text-content">
<p>Copyright © 2024 Future Data Lab,
- Design: <a rel="nofollow noopener" href="https://templatemo.com">TemplateMo</a></p>
</div>
</div>
<div class="col-lg-6 col-xs-12">
<a href="https://github.com/benjaminnNgo/ScalingTGNs/issues"> contact us</a>
</div>
</div>
</div>
</footer>
<!-- jQuery -->
<script src="assets/js/jquery-2.1.0.min.js"></script>
<!-- Bootstrap -->
<script src="assets/js/popper.js"></script>
<script src="assets/js/bootstrap.min.js"></script>
<!-- Plugins -->
<script src="assets/js/owl-carousel.js"></script>
<script src="assets/js/scrollreveal.min.js"></script>
<script src="assets/js/waypoints.min.js"></script>
<script src="assets/js/jquery.counterup.min.js"></script>
<script src="assets/js/imgfix.min.js"></script>
<script src="assets/js/slick.js"></script>
<script src="assets/js/lightbox.js"></script>
<script src="assets/js/isotope.js"></script>
<!-- Global Init -->
<script src="assets/js/custom.js"></script>
<script>
$(function() {
var selectedClass = "";
$("p").click(function(){
selectedClass = $(this).attr("data-rel");
$("#portfolio").fadeTo(50, 0.1);
$("#portfolio div").not("."+selectedClass).fadeOut();
setTimeout(function() {
$("."+selectedClass).fadeIn();
$("#portfolio").fadeTo(50, 1);
}, 500);
});
});
</script>
</body>
</html>