Skip to content

Commit a333d96

Browse files
marshkamarshka
committed
Minor fixes LoG tutorial page
1 parent d175bab commit a333d96

1 file changed

Lines changed: 67 additions & 43 deletions

File tree

‎_tutorials/graph-based-processing/log-2024.html‎

Lines changed: 67 additions & 43 deletions
Original file line numberDiff line numberDiff line change
@@ -14,11 +14,14 @@
1414
<div class="col-lg-8">
1515
<p>
1616
Successful applications of <strong>deep learning</strong> in <strong>time series</strong> processing
17-
often involve training a single neural network on a collection of (related) time series.
18-
Pairwise relationships among time series can be modeled by considering a (possibly dynamic) graph spanning the collection. In this context, <strong>graph-based
19-
methods</strong> take the standard deep learning approach to time series processing a step forward.
20-
The recent theoretical and practical developments in graph machine learning make adopting such an approach particularly appealing and timely.
21-
17+
often involve training a single neural network on a collection of (related) time series.
18+
Pairwise relationships among time series can be modeled by considering a (possibly dynamic) graph
19+
spanning the collection. In this context, <strong>graph-based
20+
methods</strong> take the standard deep learning approach to time series processing a step
21+
forward.
22+
The recent theoretical and practical developments in graph machine learning make adopting such an
23+
approach particularly appealing and timely.
24+
2225
<!-- Successful applications of <strong>deep learning</strong> in <strong>time series</strong> processing
2326
often involve training a single neural network on a collection of (related) time series. In the case
2427
of <strong>spatially correlated</strong> time series, pairwise relationships can be modeled by
@@ -33,47 +36,58 @@
3336

3437
The twofold <strong>objective</strong> of this tutorial is to:
3538
<ol>
36-
<li>offer a <strong>comprehensive overview</strong> of the field, with a focus on forecasting applications;
39+
<li>offer a <strong>comprehensive overview</strong> of the field, with a focus on forecasting
40+
applications;
3741
</li>
38-
<li>provide <strong>tools and guidelines</strong> to design and evaluate graph-based models for time series.
42+
<li>provide <strong>tools and guidelines</strong> to design and evaluate graph-based models for time
43+
series.
3944
</li>
4045
</ol>
41-
This tutorial is meant for early-carrer researchers and practitioners who wish to apply graph deep learning methods to their time series processing applications.
42-
At the same time, the tutorial provides experienced scholars with a coherent framing of the state of the art and new perspectives.
46+
This tutorial is meant for early-carrer researchers and practitioners who wish to apply graph deep
47+
learning methods to their time series processing applications.
48+
At the same time, the tutorial provides experienced scholars with a coherent framing of the state of the
49+
art and new perspectives.
4350
</p>
4451
</div>
4552
<div class="col-lg-8 text-center">
4653
<h1>Material</h1>
4754
<p>Download the slides used in our tutorial.</p>
48-
<a class="btn btn-primary" href="#" role="button">Download slides (Conf.)</a>
49-
<a class="btn btn-primary" href="#" role="button">Download slides (Meetup)</a>
55+
<div class="row g-4 justify-content-center">
56+
<div class="col-md-6">
57+
<a class="w-75 btn btn-primary" href="#" role="button">Download slides (Conf.)</a>
58+
</div>
59+
<div class="col-md-6">
60+
<a class="w-75 btn btn-primary" href="#" role="button">Download slides (Meetup)</a>
61+
</div>
62+
</div>
5063
</div>
5164
</div>
52-
</div>
5365
</section>
5466
<!-- Location section-->
55-
<section class="text-white" style="background-image: url(./img/siena-background.jpg); background-attachment: fixed; background-size: cover;">
56-
<div class="container px-4">
57-
<div class="row gx-4 justify-content-center">
58-
<div class="col-lg-8">
59-
<p class="lead">This tutorial will be delivered at the <strong
60-
class="text-decoration-underline">Learning on Graphs (LoG) Conference</strong>,
61-
held online <strong>from the 26th to the 29th of November 2024</strong>, and at the
62-
<strong class="text-decoration-underline">Italy Meetup</strong>, in Siena, Italy,
63-
<strong>from the 4th to the 6th of December 2024</strong>.<br>The tutorial will
64-
take place on
65-
<strong class="text-decoration-underline">Thursday, 28th of November</strong>, 17:00-20:00 (London, UTC+1), and
66-
<strong class="text-decoration-underline">Friday, 6th of December</strong>, 10:30-12:00 (Rome, UTC+2).
67-
</p>
68-
<div class="d-flex flex-column flex-md-row justify-content-between align-items-center"
69-
style="gap: 1.5em">
70-
<a class="btn btn-outline-light" href="https://logconference.org/" role="button"
71-
target="_blank">Conference website</a>
72-
<div>
73-
<img src="./img/log-logo.png" height="182px" class="me-2" />
67+
<section class="text-white py-0"
68+
style="background-image: url(./img/siena-background.jpg); background-attachment: fixed; background-size: cover;">
69+
<div class="w-100 h-100 py-5" style="backdrop-filter: brightness(0.5) blur(3px);">
70+
<div class="container px-4">
71+
<div class="row gx-4 justify-content-center">
72+
<div class="col-lg-8">
73+
<p class="lead">
74+
This tutorial will be delivered at the <strong class="text-decoration-underline">Learning on
75+
Graphs (LoG) Conference</strong>, held online <strong>from the 26th to the 29th of November
76+
2024</strong>, and at the <strong class="text-decoration-underline">Italy Meetup</strong>,
77+
in Siena, Italy, <strong>from the 4th to the 6th of December 2024</strong>.
78+
</p>
79+
<p class="lead">The tutorial will take place on <a class="fw-normal text-white text-decoration-underline" href="https://calendar.google.com/calendar/render?action=TEMPLATE&text=[LoG 2024] Graph Deep Learning for Time Series Processing (Tutorial)&dates=20241128T170000Z/20241128T200000Z&details=Tutorial website: https://gmlg.ch/tutorials/graph-based-processing/log-2024&location=https://logconference.org/">Thursday, 28th of November</a>, 17:00-20:00 (London, UTC), and <a class="fw-normal text-white text-decoration-underline" href="https://calendar.google.com/calendar/render?action=TEMPLATE&text=[LoG 2024 - Italy Meetup] Graph Deep Learning for Time Series Processing (Tutorial)&dates=20241206T093000Z/20241206T110000Z&details=Tutorial website: https://gmlg.ch/tutorials/graph-based-processing/log-2024&location=Via Roma, 56, 53100 Siena SI, Italia">Friday, 6th of December</a>, 10:30-12:00 (Rome, UTC+1).
80+
</p>
81+
<div class="d-flex flex-column flex-md-row justify-content-between align-items-center"
82+
style="gap: 1.5em">
83+
<a class="btn btn-outline-light" href="https://logconference.org/" role="button"
84+
target="_blank">Conference website</a>
85+
<div>
86+
<img src="./img/log-logo.png" height="182px" class="me-2" />
87+
</div>
88+
<a class="btn btn-outline-light" href="https://sites.google.com/student.unisi.it/log24siena"
89+
role="button" target="_blank">Meetup website</a>
7490
</div>
75-
<a class="btn btn-outline-light" href="https://sites.google.com/student.unisi.it/log24siena" role="button"
76-
target="_blank">Meetup website</a>
7791
</div>
7892
</div>
7993
</div>
@@ -85,39 +99,49 @@ <h1>Material</h1>
8599
<div class="row gx-4 justify-content-center">
86100
<div class="col-lg-8">
87101
<h1 class="text-center">Program</h1>
88-
<h5 class="mt-3 mb-2 fw-light"><span class="fw-bold me-2">Part 1</span> Graph deep learning for time series processing.</h5>
102+
<h5 class="mt-3 mb-2 fw-light"><span class="fw-bold me-2">Part 1</span> Graph deep learning for time
103+
series processing.</h5>
89104
<ol>
90105
<li><strong>Correlated time series</strong><br>
91106
Definition of the problem settings. Time series forecasting.
92107
</li>
93108
<li><strong>Graph deep learning for time series forecasting</strong><br>
94-
Graph-based framework for representing correlated time series. Graph-based models for time series forecasting.
95-
Similarities to and differences from related settings in time series analysis and temporal graph learning.
109+
Graph-based framework for representing correlated time series. Graph-based models for time
110+
series forecasting.
111+
Similarities to and differences from related settings in time series analysis and temporal graph
112+
learning.
96113
</li>
97114
<li><strong>Spatiotemporal graph neural networks (STGNNs)</strong><br>
98-
Core components of a STGNN model. Recipes and strategies for building STGNNs. The time-then-space and time-and-space paradigms.
115+
Core components of a STGNN model. Recipes and strategies for building STGNNs. The
116+
time-then-space and time-and-space paradigms.
99117
Overview of architectures from the literature.
100118
</li>
101119
<li><strong>Global and local models</strong><br>
102-
Parameter sharing in time series models. Review of the global and local modeling paradigms with their strengths.
103-
Practical implications in graph-based time series processing. Hybrid global-local STGNN architectures. Transfer learning.
120+
Parameter sharing in time series models. Review of the global and local modeling paradigms with
121+
their strengths.
122+
Practical implications in graph-based time series processing. Hybrid global-local STGNN
123+
architectures. Transfer learning.
104124
</li>
105125
</ol>
106126
<h5 class="mt-3 mb-2 fw-light"><span class="fw-bold me-2">Demo</span> Coding Spatiotemporal GNNs</h5>
107-
<p>Overview of open-source Pytorch libraries for graph-based time series processing. Torch Spatiotemporal demo.</p>
127+
<p>Overview of open-source Pytorch libraries for graph-based time series processing. Torch
128+
Spatiotemporal demo.</p>
108129
<h5 class="mt-3 mb-2 fw-light"><span class="fw-bold me-2">Part 2</span> Challenges and tools.</h5>
109130
<ol>
110131
<li><strong>Latent graph learning</strong><br>
111-
Methods to apply the framework when no pre-defined graph is available. Learning graph structures from collections of time series.
132+
Methods to apply the framework when no pre-defined graph is available. Learning graph structures
133+
from collections of time series.
112134
</li>
113135
<li><strong>Scalability</strong><br>
114136
Methods to enable scalability to large sensor networks.
115137
</li>
116138
<li><strong>Dealing with missing data</strong><br>
117-
The problem of missing data. Overview of methods for graph-based multivariate time series imputation and kriging.
139+
The problem of missing data. Overview of methods for graph-based multivariate time series
140+
imputation and kriging.
118141
</li>
119142
<li><strong>Model quality assessment</strong><br>
120-
Statistical tools to test the optimality of graph-based predictors. Identification of time-space regions where predictions can be improved.
143+
Statistical tools to test the optimality of graph-based predictors. Identification of time-space
144+
regions where predictions can be improved.
121145
</li>
122146
</ol>
123147
</div>

0 commit comments

Comments
 (0)