-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathindex.html
708 lines (642 loc) · 27.3 KB
/
index.html
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
<!DOCTYPE html>
<html lang="en">
<head>
<title>CS839 - Modern Data Management and Machine Learning Systems (Spring 2020)</title>
<!-- HTML5 Shim and Respond.js IE8 support of HTML5 elements and media queries -->
<!-- WARNING: Respond.js doesn't work if you view the page via file:// -->
<!--[if lt IE 9]>
<script src="https://oss.maxcdn.com/html5shiv/3.7.2/html5shiv.min.js"></script>
<script src="https://oss.maxcdn.com/respond/1.4.2/respond.min.js"></script>
<![endif]-->
<link rel="stylesheet" href="https://maxcdn.bootstrapcdn.com/bootstrap/3.2.0/css/bootstrap.min.css">
<script src="https://ajax.googleapis.com/ajax/libs/jquery/2.1.1/jquery.min.js"></script>
<script src="https://maxcdn.bootstrapcdn.com/bootstrap/3.2.0/js/bootstrap.min.js"></script>
<link rel="stylesheet" href="./style.css">
</head>
<div class="container">
<div class="navbar navbar-default" role="navigation">
<div class="container-fluid">
<div class="navbar-header">
<button type="button" class="navbar-toggle collapsed" data-toggle="collapse" data-target=".navbar-collapse">
<span class="sr-only">Toggle navigation</span>
<span class="icon-bar"></span>
<span class="icon-bar"></span>
<span class="icon-bar"></span>
</button>
<a class="navbar-brand" href="#">CS 839</a>
</div>
<div class="collapse navbar-collapse" id="bs-example-navbar-collapse-1">
<ul class="nav navbar-nav">
<li class="dropdown">
<a href="#" class="dropdown-toggle" data-toggle="dropdown">About <span class="caret"></span></a>
<ul class="dropdown-menu" role="menu">
<li><a href="#description">Description</a></li>
<li><a href="#logistics">Class Logistics</a></li>
<li><a href="#lecture">Lecture Plan</a></li>
<li><a href="#grading">Grading</a></li>
<li class="divider"></li>
<li><a href="#late-policy">Late Policy</a></li>
</ul>
</li>
<li><a href="#office-hours">Office Hours</a></li>
</ul>
</div>
</div>
</div>
<div class="jumbotron">
<h1>CS839 Modern Data Management and Machine Learning Systems</h1>
<h3>COMP SCI 1325 on TuTh 9:30AM - 10:45AM</h3>
<p class="lead"> Class mailing list <a href="mailto:[email protected]">[email protected]</a>. We will be using it for announcements.</p>
<p class="lead"> <font color=red><b>Instructions for online classes can be found</b> <a href="online.txt">here</a></font>. </p>
</div>
<div class="panel panel-default">
<div class="panel-heading">Description</div>
<div class="panel-body">
<a name="description"></a>
<p> Modern Data Analytics and Machine learning (ML) are enjoying rapidly increasing adoption. However, designing and implementing the systems that support modern data analytics and machine learning in real-world deployments presents a significant challenge, in large part due to the radically different development and deployment profile of modern data analysis methods, and the range of practical concerns that come with broader adoption.</p>
<p> In this seminar course, we will describe the latest trends in modern data management and machine learning systems designs to better support the next generation of ML applications, and applications of ML to optimize the architecture and the performance of data management systems.</p>
</div>
</div>
<div class="panel panel-default">
<div class="panel-heading">Class Logistics</div>
<div class="panel-body">
<a name="logistics"></a>
<p> <b>Class Format</b>
<ul>
<li><p>This is a seminar course. The format of this course will be a mix of lectures, seminar-style discussions, student presentations. Students will be responsible for paper readings, conference-style reviews, and completing a hands-on project. For projects, we will strongly encourage teams of <b>three people</b>.</p></li>
</ul>
</p>
<p><b>Assignments</b>
<ul>
<li>Paper reviews/presentations (Individual assignments).
<li>Semester Project (Group assignment). We encourage students to find projects that relate to their ongoing research.</li>
<li>There will be <b> no midterm or final exams</b>.</li>
</ul>
</p>
<p><b>Misc</b>
<ul>
<li>The lecture schedule may be updated to accomodate obligations of the instructor.</li>
</ul>
</p>
</div>
</div>
<div class="panel panel-default">
<div class="panel-heading">
<h2 class="panel-title">Tentative Lecture Plan (Subject to Change)</h2>
</div>
<div class="panel-body">
<a name="lecture"></a>
<p>
<hr />
<table class="plan">
<tbody>
<tr>
<th>Week</th>
<th>Date</th>
<th>Topic</th>
<th>Lecture Materials</th>
<th>Reading Material</th>
<th>Assignments</th>
</tr>
<tr>
<td align="center" colspan="6" class="small-section-header">Introduction and Class Overview</td>
</tr>
<tr>
<td>1</td>
<td>1/21</td>
<td>Logistics and The Technical Aspects of Production ML</td>
<td class="side-borders">
<!-- Lecture -->
<a href="lectures/Lecture_0_Intro.pptx">Lecture 1</a> <a href="lectures/Lecture_0_Intro.pdf">(pdf)</a>
</td>
<td class="side-borders">
<ul>
<li>MLSys: <a href="https://arxiv.org/abs/1904.03257">MLSys Whitepaper</a></li>
<li><a href="https://homes.cs.washington.edu/~pedrod/papers/cacm12.pdf">A Few Useful Things to Know About Machine Learning</a></li>
<li>Polyzotis et al. <a href="https://research.google/pubs/pub46178/">Data Management Challenges in Production Machine Learning</a>, SIGMOD 2017</li>
</ul>
</td>
<td>
<li>Sign-up for paper presentations and scribes <a href="https://docs.google.com/spreadsheets/d/1VbO7NloceXI4AU4vk0QwwTLbWUgb8i5kiMlnhSk6H-0/edit?usp=sharing">here</a>.</li>
</td>
</tr>
<tr>
<td align="center" colspan="6" class="small-section-header">Machine Learning Life-Cycle</td>
</tr>
<tr>
<td>1</td>
<td>1/23</td>
<td>Machine Learning Life-Cycle: A systems' percpective</td>
<td class="side-borders">
<!-- Lecture -->
<a href="lectures/Lecture_1_ML_lifecycle.pdf">Lecture 2 (pdf)</a>
</td>
<td class="side-borders"></td>
<td></td>
</tr>
<tr>
<td>2</td>
<td>1/28</td>
<td>Machine Learning Life-Cycle: A systems' percpective</td>
<td class="side-borders">
<!-- Lecture -->
<a href="">Paper Review and Discussion</a>
</td>
<td class="side-borders">
<ul>
<li><a href="https://papers.nips.cc/paper/5656-hidden-technical-debt-in-machine-learning-systems.pdf">Hidden Technical Debt in Machine Learning Systems</a> (to review)</li>
<li><a href="http://sites.computer.org/debull/A18dec/issue1.htm">Data Engineering Bulletin: Machine Learning Life-cycle Management</a></li>
<li><a href="http://sites.computer.org/debull/A18dec/p39.pdf">Accelerating the Machine Learning Lifecycle with MLflow</a></li>
<li><a href="https://research.google/pubs/pub46484/">TFX: A TensorFlow-Based Production-Scale Machine Learning Platform</a></li>
<li><a href="https://www-cs.stanford.edu/~matei/papers/2016/hilda_modeldb.pdf">MODELDB: A System for Machine Learning Model Management</a></li>
</ul>
</td>
<td>
<li>Submit your reviews in Canvas.</li>
</td>
</tr>
<tr>
<td>2</td>
<td>1/30</td>
<td><b>No Class</b></td>
<td class="side-borders">
<!-- Lecture -->
<b>No Lecture</b>
</td>
<td class="side-borders"></td>
<td></td>
</tr>
<tr>
<td align="center" colspan="6" class="small-section-header">Training Data Collection</td>
</tr>
<tr>
<td>3</td>
<td>2/4</td>
<td>Weak-Supervision: Automating The Collection of Training Data</td>
<td class="side-borders">
<!-- Lecture -->
<a href="lectures/Lecture_2_Training_Data_Collection.pdf">Lecture 3 (pdf)</a>
</td>
<td class="side-borders">
<ul>
<li><a href="http://cs229.stanford.edu/notes2019fall/weak_supervision_notes.pdf">Weak Supervision Notes</a></li>
<li><a href-"http://www.offconvex.org/2019/03/19/CURL/">Notes on Self-Supervised Learning</a></li>
<li><a href-"https://arxiv.org/abs/1706.05098">Multi-Task Learning Overview</a></li>
</ul>
</td>
<td>
</td>
</tr>
<tr>
<td>3</td>
<td>2/6</td>
<td>Weak-Supervision: Automating The Collection of Training Data</td>
<td class="side-borders">
<!-- Lecture -->
<a href="">Paper Review and Discussion</a>
</td>
<td class="side-borders">
<ul>
<li><a href="https://arxiv.org/abs/1810.02840">Training Complex Models with Multi-Task Weak Supervision
</a> (to review)</li>
<li><a href="https://arxiv.org/abs/1902.09229">A Theoretical Analysis of Contrastive Unsupervised Representation Learning</a></li>
<li><a href="https://arxiv.org/abs/1909.07930">Ludwig: a type-based declarative deep learning toolbox
</a></li>
<li><a href="https://arxiv.org/pdf/1811.03402">A Survey on Data Collection for Machine Learning: A Big Data - AI Integration Perspective</a></li>
<li><a href="http://openaccess.thecvf.com/content_CVPR_2019/papers/Cubuk_AutoAugment_Learning_Augmentation_Strategies_From_Data_CVPR_2019_paper.pdf">AutoAugment: Learning Augmentation Strategies From Data</a></li>
</ul>
</td>
<td><li>Submit your reviews in Canvas.</li></td>
</tr>
<tr>
<td align="center" colspan="6" class="small-section-header">Data Validation</td>
</tr>
<tr>
<td>4</td>
<td>2/11</td>
<td>Managing Noisy Data</td>
<td class="side-borders">
<!-- Lecture -->
<p><a href="lectures/Lecture_3_Data_Validation.pdf">Lecture 4 (pdf)</a></p>
<p><a href="http://pages.cs.wisc.edu/~thodrek/MLND.pdf">Additional Slides</a></p>
</td>
<td class="side-borders"></td>
<td>
</td>
</tr>
<tr>
<td>4</td>
<td>2/13</td>
<td>Managing Noisy Data</td>
<td class="side-borders">
<!-- Lecture -->
<a href="">Paper Review and Discussion</a>
</td>
<td class="side-borders">
<ul>
<li><a href="https://arxiv.org/abs/1801.06750">A Formal Framework For Probabilistic Unclean Databases</a> (to review)</li>
<li><a href="https://arxiv.org/pdf/1811.00741.pdf">Stronger Data Poisoning Attacks Break Data Sanitization Defenses</a></li>
<li><a href="http://proceedings.mlr.press/v37/globerson15.html">How Hard is Inference for Structured Prediction?</a></li>
<li><a href="http://proceedings.mlr.press/v97/diakonikolas19a/diakonikolas19a.pdf">Sever: A Robust Meta-Algorithm for Stochastic Optimization</a></li>
<li><a href="https://openreview.net/forum?id=rJx1Na4Fwr">MACER: Attack-free and Scalable Robust Training via Maximizing Certified Radius</a></li>
</ul>
</td>
<td><li>Submit your reviews in Canvas.</li></td>
</tr>
<tr>
<td align="center" colspan="6" class="small-section-header">Adaptive Data Management</td>
</tr>
<tr>
<td>5</td>
<td>2/18</td>
<td>Learnable Data Structures</td>
<td class="side-borders">
<!-- Lecture -->
<a href="lectures/Lecture_4_Adaptive_Data_Structures.pdf">Lecture 5 (pdf)</a>
</td>
<td class="side-borders"></td>
<td><li>Register your Project Groups <a href="https://docs.google.com/spreadsheets/d/1g55_2LjU4AHLN0ryfwrixRTb3be6k-aAmmDHSx96Ks8/edit?usp=sharing">here</a>.</li>
</td>
</tr>
<tr>
<td>5</td>
<td>2/20</td>
<td>Learnable Data Structures</td>
<td class="side-borders">
<!-- Lecture -->
<a href="">Paper Review and Discussion</a>
</td>
<td class="side-borders">
<ul>
<li><a href="https://www.cl.cam.ac.uk/~ey204/teaching/ACS/R244_2018_2019/papers/Kraska_SIGMOD_2018.pdf">The Case for Learned Index Structures</a> (to review)</li>
<li><a href="https://stratos.seas.harvard.edu/files/stratos/files/learningdatastructurealchemy.pdf">Learning Data Structure Alchemy</a></li>
<li><a href="https://arxiv.org/pdf/1912.01668">Learning Multi-Dimensional Indexes</a></li>
<li><a href="https://stratos.seas.harvard.edu/files/stratos/files/periodictabledatastructures.pdf">The Periodic Table of Data Structures</a></li>
<li><a href="https://arxiv.org/abs/1809.00677">Learned Cardinalities: Estimating Correlated Joins with Deep Learning</a></li>
</ul>
</td>
<td><li>Submit your reviews in Canvas.</li></td>
</tr>
<tr>
<td align="center" colspan="6" class="small-section-header">Project Proposal Review</td>
</tr>
<tr>
<td>6</td>
<td>2/25</td>
<td><b>No Class</b></td>
<td class="side-borders">
<!-- Lecture -->
<a href="">Proposal Submission Deadline (by end-of-day).</a>
</td>
<td class="side-borders">
</td>
<td>
<li>Please read the instructions for submitting your Project Proposal on Canvas.</li>
</td>
</tr>
<tr>
<td>6</td>
<td>2/27</td>
<td><b>Meetings with Instructor to Review Project Proposals</b></td>
<td class="side-borders">
<!-- Lecture -->
</td>
<td class="side-borders">
<ul>
<li>Sign-up sheet (15-minute slots):</li>
</ul>
</td>
<td></td>
</tr>
<tr>
<td align="center" colspan="6" class="small-section-header">Project Proposal Review</td>
</tr>
<tr>
<td>7</td>
<td>3/3</td>
<td><b>No Class (MLSys 2020)</b></td>
<td class="side-borders">
<!-- Lecture -->
<b> No Lecture </b>
</td>
<td class="side-borders"></td>
<td>
</td>
</tr>
<tr>
<td>7</td>
<td>3/5</td>
<td><b>Meetings with Instructor to Review Project Proposals</b></td>
<td class="side-borders">
<!-- Lecture -->
</td>
<td class="side-borders">
<ul>
<li>Sign-up sheet (15-minute slots):</li>
</ul>
</td>
<td></td>
</tr>
<tr>
<td align="center" colspan="6" class="small-section-header">ML in Database Systems</td>
</tr>
<tr>
<td>8</td>
<td>3/10</td>
<td>ML in Database Systems</td>
<td class="side-borders">
<!-- Lecture -->
<a href="lectures/Lecture_5_ML_in_DB.pdf">Lecture 6 (pdf)</a>
</td>
<td class="side-borders"></td>
<td>
</td>
</tr>
<tr>
<td>8</td>
<td>3/12</td>
<td>Optimizations for Feature Selection Workloads</td>
<td class="side-borders">
<!-- Lecture -->
<a href="">Paper Review and Discussion</a>
</td>
<td class="side-borders">
<ul>
<li><a href="https://dl.acm.org/citation.cfm?id=2877204">Materialization optimizations for feature selection workloads</a> (to review)</li>
<li><a href="https://adalabucsd.github.io/papers/2015_Orion_SIGMOD.pdf">Learning generalized linear models over normalized data</a></li>
<li><a href="https://sigmodrecord.org/publications/sigmodRecord/1606/pdfs/03_principles_Olteanu.pdf">Factorized Databases</a></li>
<li><a href="http://www.cs.ox.ac.uk/dan.olteanu/papers/nnos-amw17.pdf">In-Database Factorized Learning</a></li>
<li><a href="https://ora.ox.ac.uk/objects/uuid:2d852e0d-889d-46fe-890e-b1ac5687c798/download_file?file_format=pdf&safe_filename=main.pdf&type_of_work=Conference+item">In-Database Learning with Sparse Tensors</a></li>
</ul>
</td>
<td><li>Submit your reviews in Canvas.</li></td>
</tr>
<tr>
<td align="center" colspan="6"><font color=red><b>Spring Break</b></font></td>
</tr>
<tr>
<td align="center" colspan="6" class="small-section-header">Automated Machine Learning (AutoML)</td>
</tr>
<tr>
<td>10</td>
<td>3/24</td>
<td>Efficient and Robust Automated Machine Learning</td>
<td class="side-borders">
<!-- Lecture -->
<a href="lectures/Lecture_6_AutoML.pdf">Lecture 7 (pdf)</a>
</td>
<td class="side-borders">
<ul>
<li><a href="lectures/lecture-automl-Mar-24-2020.pdf">Lecture notes</a></li>
<li><a href="https://www.cs.cornell.edu/courses/cs4787/2019sp/notes/lecture16.pdf">Bayesian optimization lecture notes</a></li>
<li><a href="https://zoom.us/rec/share/_uBzLbz16UZIGbfIzBr-VbIrI47laaa81yIZqfNemo_Wdez0_wSyzXGu7BaJNDs">Recorded lecture</a></li>
</ul>
</td>
<td>
</td>
</tr>
<tr>
<td>10</td>
<td>3/26</td>
<td>Efficient and Robust Automated Machine Learning</td>
<td class="side-borders">
<!-- Lecture -->
<a href="">Paper Review and Discussion</a>
</td>
<td class="side-borders">
<p><b>Lecture Notes</b></p>
<ul>
<li><a href="lectures/lecture-automl-2.pdf">Lecture notes</a></li>
<li><a href="https://zoom.us/rec/share/xMwyHb758WFJGaeRr0_-B4RmAKe1eaa80CBNq_JfzRmX816vzlzjqgfjeb1FTL5U">Recorded lecture</a></li>
</ul>
<p><b>Reading</b></p>
<ul>
<li><a href="https://papers.nips.cc/paper/5872-efficient-and-robust-automated-machine-learning.pdf">Efficient and Robust Automated Machine Learning</a> (to review)</li>
<li><a href="https://research.google/pubs/pub45826/">Neural Architecture Search with Reinforcement Learning</a></li>
<li><a href="https://papers.nips.cc/paper/4522-practical-bayesian-optimization-of-machine-learning-algorithms.pdf">Practical Bayesian Optimization of Machine Learning Algorithms</a></li>
<li><a href="https://dl.acm.org/citation.cfm?id=3319863">Democratizing Data Science through Interactive Curation of ML Pipelines</a></li>
<li><a href="http://proceedings.mlr.press/v97/so19a/so19a.pdf">The Evolved Transformer</a></li>
</ul>
</td>
<td><li>Submit your reviews in Canvas.</li></td>
</tr>
<tr>
<td align="center" colspan="6" class="small-section-header">Distributed Model Training</td>
</tr>
<tr>
<td>11</td>
<td>3/31</td>
<td>Distributed Model Training</td>
<td class="side-borders">
<!-- Lecture -->
<a href="lectures/Lecture_7_Distributed_ML.pdf">Lecture 8 (pdf)</a>
</td>
<td class="side-borders">
<ul>
<li><a href="https://zoom.us/rec/share/tcd0AKjcyTpLbYmXtFnESKgMOpvAT6a80ygfqKcFn0lxtcWcmznumlCG2G-PtShg?startTime=1585665369000">Recorded lecture (part 1)</a></li>
<li><a href="https://zoom.us/rec/share/tcd0AKjcyTpLbYmXtFnESKgMOpvAT6a80ygfqKcFn0lxtcWcmznumlCG2G-PtShg?startTime=1585668130000">Recorded lecture (part 2)</a></li>
</ul>
</td>
<td>
</td>
</tr>
<tr>
<td>11</td>
<td>4/2</td>
<td>Distributed Model Training</td>
<td class="side-borders">
<!-- Lecture -->
<a href="">Paper Review and Discussion</a>
</td>
<td class="side-borders">
<ul>
<li><a href="https://www.usenix.org/system/files/conference/osdi14/osdi14-paper-li_mu.pdf">Scaling Distributed Machine Learning with the Parameter Server</a> (to review)</li>
<li><a herf="https://people.eecs.berkeley.edu/~brecht/papers/hogwildTR.pdf">Hogwild!: A Lock-Free Approach to Parallelizing Stochastic Gradient Descent</a></li>
<li><a href="https://static.googleusercontent.com/media/research.google.com/en//archive/large_deep_networks_nips2012.pdf">Large Scale Distributed Deep Networks</a></li>
<li><a href="https://arxiv.org/abs/1806.04090">ATOMO: Communication-efficient Learning via Atomic Sparsification</a></li>
<li><a href="https://arxiv.org/pdf/1903.06701.pdf">Scaling Distributed Machine Learning with In-Network Aggregation</a></li>
</ul>
</td>
<td><li>Submit your reviews in Canvas.</li></td>
</tr>
<tr>
<td align="center" colspan="6" class="small-section-header">Efficient Prediction Serving</td>
</tr>
<tr>
<td>12</td>
<td>4/7</td>
<td>Model Cascades</td>
<td class="side-borders">
<!-- Lecture -->
<a href="lectures/Lecture_8_Efficient_model_serving.pdf">Lecture 9 (pdf)</a>
</td>
<td class="side-borders"><a href="lectures/lecture-notes-model-serving.pdf">Lecture notes(pdf)</a></td>
<td>
</td>
</tr>
<tr>
<td>12</td>
<td>4/9</td>
<td>Model Cascades</td>
<td class="side-borders">
<!-- Lecture -->
<a href="">Paper Review and Discussion</a>
</td>
<td class="side-borders">
<ul>
<li><a href="https://www.usenix.org/conference/nsdi17/technical-sessions/presentation/crankshaw">Clipper: A Low-Latency Online Prediction Serving System</a> (to review)</li>
<li><a href="https://arxiv.org/abs/1906.01974">Willump: A Statistically-Aware End-to-end Optimizer for Machine Learning Inference</a></li>
<li><a href="https://www.usenix.org/system/files/osdi18-hsieh.pdf">Focus: Querying Large Video Datasets with Low Latency and Low Cost</a></li>
<li><a href="https://www.usenix.org/system/files/osdi18-lee.pdf">Pretzel: Opening the Black Box of Machine Learning Prediction Serving Systems</a></li>
<li><a href="https://www.microsoft.com/en-us/research/uploads/prod/2017/12/s18_cr3.pdf">Accelerating Machine Learning Inference with Probabilistic Predicates</a></li>
</ul>
</td>
<td><li>Submit your reviews in Canvas.</li></td>
</tr>
<tr>
<td align="center" colspan="6" class="small-section-header">Model Compression</td>
</tr>
<tr>
<td>13</td>
<td>4/14</td>
<td>Low-Precision and Low-Rank Learning</td>
<td class="side-borders">
<!-- Lecture -->
<a href="lectures/Lecture_9_Compression.pdf">Lecture 10 (pdf)</a>
</td>
<td class="side-borders"><a href="lectures/lecturenotes-compression.pdf">Lecture notes (pdf)</a></td>
<td>
</td>
</tr>
<tr>
<td>13</td>
<td>4/16</td>
<td>Low-Precision and Low-Rank Learning</td>
<td class="side-borders">
<!-- Lecture -->
<a href="">Paper Review and Discussion</a>
</td>
<td class="side-borders">
<ul>
<li><a href="https://arxiv.org/abs/1803.03635">The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks</a> (to review)</li>
<li><a href="https://arxiv.org/abs/1412.7024">Training deep neural networks with low precision multiplications</a></li>
<li><a href="https://arxiv.org/abs/1710.09282">A Survey of Model Compression and Acceleration for Deep Neural Networks</a></li>
<li><a href="http://www.cs.cornell.edu/~cdesa/papers/isca2017_buckwild.pdf">Understanding and Optimizing Asynchronous Low-Precision Stochastic Gradient Descent</a></li>
<li><a href="https://arxiv.org/abs/1503.02531">Distilling the Knowledge in a Neural Network</a></li>
</ul>
</td>
<td><li>Submit your reviews in Canvas.</li></td>
</tr>
<tr>
<td align="center" colspan="6" class="small-section-header">Relational Inductive Bias in Neural Networks</td>
</tr>
<tr>
<td>14</td>
<td>4/21</td>
<td>Graph (Neural) Networks</td>
<td class="side-borders">
<!-- Lecture -->
<a href="">Lecture 11</a>
</td>
<td class="side-borders"></td>
<td>
</td>
</tr>
<tr>
<td>14</td>
<td>4/23</td>
<td>Graph (Neural) Networks</td>
<td class="side-borders">
<!-- Lecture -->
<a href="">Paper Review and Discussion</a>
</td>
<td class="side-borders">
<ul>
<li><a href="https://arxiv.org/abs/1806.01261">Relational inductive biases, deep learning, and graph networks</a> (to review)</li>
<li><a href="https://arxiv.org/abs/1901.00596">A Comprehensive Survey on Graph Neural Networks</a></li>
<li><a href="https://arxiv.org/abs/1809.10341">Deep Graph Infomax</a></li>
<li><a href="http://proceedings.mlr.press/v80/sala18a.html">Representation Tradeoffs for Hyperbolic Embeddings</a></li>
<li><a href="https://www.aaai.org/ocs/index.php/AAAI/AAAI17/paper/viewPaper/14967">Label-Free Supervision of Neural Networks with Physics and Domain Knowledge</a></li>
</ul>
</td>
<td><li>Submit your reviews in Canvas.</li></td>
</tr>
<tr>
<td align="center" colspan="6" class="small-section-header">Project Presentations Week</td>
</tr>
<tr>
<td>15</td>
<td>4/28</td>
<td><b>Project Presentations Part 1</b></td>
<td class="side-borders">
<!-- Lecture -->
</td>
<td class="side-borders"></td>
<td>
</td>
</tr>
<tr>
<td>15</td>
<td>4/30</td>
<td><b>Project Presentations Part 2</b></td>
<td class="side-borders">
<!-- Lecture -->
</td>
<td class="side-borders">
</td>
<td></td>
</tr>
<tr>
<td align="center" colspan="6" class="small-section-header">Final Project Report</td>
</tr>
<tr>
<td>16</td>
<td>5/8</td>
<td><b>Final Project Report Due (by end-of-day)</b></td>
<td class="side-borders">
<!-- Lecture -->
</td>
<td class="side-borders"></td>
<td>
<li>Instructions for submitting your Final Project Report can be found <a href="">here</a>.</li>
</td>
</tr>
</tbody>
</table>
</p>
</div>
</div>
<div class="panel panel-default">
<div class="panel-heading">Grading</div>
<div class="panel-body">
<a name="grading"></a>
<table class="table">
<tr><td>Reviews and Presentations</td><td>20%</td>
<tr><td>Project Presentation</td><td>20%</td>
<tr><td>Project Final Report</td><td>60%</td>
</table>
</div>
</div>
<div class="panel panel-default">
<div class="panel-heading">Office Hours</div>
<div class="panel-body">
<a name="office-hours"></a>
<p>Theo: by appointment @ Room CS4361</p>
</div>
</div>
<div class="panel panel-default">
<div class="panel-heading">Late Policy and Deliverables</div>
<div class="panel-body">
<a name="late-policy"></a>
There will be <b>no late dates for the project deliverables</b>. However, you have the option to <b>skip up to two reviews</b>. Additional extensions may be granted in the case of a severe medical or family emergency.
</div>
</div>
<div class="panel panel-default">
<div class="panel-heading">Credit</div>
<div class="panel-body">
The template of this website was created by HazyReseach@Stanford.
</div>
</div>
</div>
</body>
</html>