Updated tutorial website

Author: andreacini

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

@@ -1,6 +1,6 @@
 ---
 layout: particles_header
-title: Graph Deep Learning for Spatiotemporal Time Series
+title: Graph Deep Learning for Time Series Processing
 lead: Forecasting, Reconstruction and Analysis
 description: The GMLG tutorial on graph deep learning for time-series processing.
 venue: LoG Conference 2024
@@ -14,13 +14,11 @@
             <div class="col-lg-8">
                 <p>
                     Successful applications of <strong>deep learning</strong> in <strong>time series</strong> processing
-                    often involve training a single neural network on a collection of (related) time series. In the case
-                    of <strong>spatially correlated</strong> time series, pairwise relationships can be modeled by
-                    considering a (possibly dynamic) graph spanning the collection. In this context, <strong>graph-based
-                        methods</strong> take the standard deep learning approach to time series processing a step
-                    forward. The recent theoretical and practical developments in graph neural networks make the
-                    adoption of such an approach particularly appealing and timely.
-
+                    often involve training a single neural network on a collection of (related) time series. 
+                    Pairwise relationships among time series can be modeled by considering a (possibly dynamic) graph spanning the collection. In this context, <strong>graph-based
+                        methods</strong> take the standard deep learning approach to time series processing a step forward. 
+                    The recent theoretical and practical developments in graph machine learning make adopting such an approach particularly appealing and timely. 
+    
                     <!-- Successful applications of <strong>deep learning</strong> in <strong>time series</strong> processing
                     often involve training a single neural network on a collection of (related) time series. In the case
                     of <strong>spatially correlated</strong> time series, pairwise relationships can be modeled by
@@ -35,17 +33,13 @@
 
                     The twofold <strong>objective</strong> of this tutorial is to:
                 <ol>
-                    <li>offer a <strong>comprehensive overview</strong> of the field, with a focus on forecasting and
-                        missing data reconstruction;
+                    <li>offer a <strong>comprehensive overview</strong> of the field, with a focus on forecasting applications;
                     </li>
-                    <li>provide the necessary theoretically grounded <strong>tools and practical
-                            recommendations</strong> to design and evaluate graph-based spatiotemporal models.
+                    <li>provide <strong>tools and guidelines</strong> to design and evaluate graph-based models for time series.
                     </li>
                 </ol>
-                This tutorial is meant for both early-career and experienced researchers looking for a coherent
-                framing of the field, seeking principles and guidelines to secure the most advanced technologies.
-                Basic knowledge about deep learning for time series and graph-structured data is sufficient to
-                benefit from the tutorial.
+                This tutorial is meant for early-carrer researchers and practitioners who wish to apply graph deep learning methods to their time series processing applications. 
+                At the same time, the tutorial provides experienced scholars with a coherent framing of the state of the art and new perspectives.
                 </p>
             </div>
             <div class="col-lg-8 text-center">
@@ -62,9 +56,9 @@ <h1>Material</h1>
     <div class="container px-4">
         <div class="row gx-4 justify-content-center">
             <div class="col-lg-8">
-                <p class="lead">This tutorial will be delivered mainly at the <strong
+                <p class="lead">This tutorial will be delivered at the <strong
                         class="text-decoration-underline">Learning on Graphs (LoG) Conference</strong>,
-                    held virtually <strong>from the 26th to the 29th of November 2024</strong>, and concludes at the 
+                    held online <strong>from the 26th to the 29th of November 2024</strong>, and at the 
                     <strong class="text-decoration-underline">Italy Meetup</strong>, in Siena, Italy, 
                     <strong>from the 4th to the 6th of December 2024</strong>.<br>The tutorial will
                     take place on 
@@ -91,52 +85,41 @@ <h1>Material</h1>
         <div class="row gx-4 justify-content-center">
             <div class="col-lg-8">
                 <h1 class="text-center">Program</h1>
-                <h5 class="mt-3 mb-2 fw-light"><span class="fw-bold me-2">Part 1</span> Graph-based processing of
-                    spatiotemporal time series.</h5>
+                <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>
                 <ol>
-                    <li><strong>Spatiotemporal time series</strong><br>
-                        Definition of the problem settings. Introduction to common downstream tasks: forecasting and
-                        imputation.
+                    <li><strong>Correlated time series</strong><br>
+                        Definition of the problem settings. Time series forecasting.
+                    </li>
+                    <li><strong>Graph deep learning for time series forecasting</strong><br>
+                        Graph-based framework for representing correlated time series. Graph-based models for time series forecasting.
+                        Similarities to and differences from related settings in time series analysis and temporal graph learning.
                     </li>
                     <li><strong>Spatiotemporal graph neural networks (STGNNs)</strong><br>
-                        Presentation of the fundamental components of the general STGNN family of deep learning models
-                        for STS. Recipes and strategies for building effective STGNNs, as well as architectures from the
-                        literature, are provided.
+                        Core components of a STGNN model. Recipes and strategies for building STGNNs. The time-then-space and time-and-space paradigms.
+                        Overview of architectures from the literature.
                     </li>
                     <li><strong>Global and local models</strong><br>
-                        Discussion on the problem of local effects in spatiotemporal data. Review of the global and
-                        local modeling paradigms with their strengths and practical implications.
-                    </li>
-                    <li><strong>Model quality assessment</strong><br>
-                        Identification of time-space regions, e.g., specific sensors or periods of time, where
-                        predictions can be improved.
+                        Parameter sharing in time series models. Review of the global and local modeling paradigms with their strengths. 
+                        Practical implications in graph-based time series processing. Hybrid global-local STGNN architectures. Transfer learning.
                     </li>
                 </ol>
-
+                <h5 class="mt-3 mb-2 fw-light"><span class="fw-bold me-2">Demo</span> Coding Spatiotemporal GNNs</h5>
+                <p>Overview of open-source Pytorch libraries for graph-based time series processing. Torch Spatiotemporal demo.</p>
                 <h5 class="mt-3 mb-2 fw-light"><span class="fw-bold me-2">Part 2</span> Challenges and tools.</h5>
                 <ol>
                     <li><strong>Latent graph learning</strong><br>
-                        Why and how to learn a graph structure from data when relational information is unavailable,
-                        insufficient or unreliable.
-                    </li>
-                    <li><strong>Learning in non-stationary environments</strong><br>
-                        Challenges and methods associated with modeling the evolution of spatiotemporal systems over
-                        time.
+                        Methods to apply the framework when no pre-defined graph is available. Learning graph structures from collections of time series.
                     </li>
                     <li><strong>Scalability</strong><br>
                         Methods to enable scalability to large sensor networks.
                     </li>
                     <li><strong>Dealing with missing data</strong><br>
-                        The problem of missing data in real-world applications and architectures for graph-based
-                        multivariate
-                        time series imputation.
+                        The problem of missing data. Overview of methods for graph-based multivariate time series imputation and kriging.
+                    </li>
+                    <li><strong>Model quality assessment</strong><br>
+                        Statistical tools to test the optimality of graph-based predictors. Identification of time-space regions where predictions can be improved.
                     </li>
                 </ol>
-
-                <h5 class="mt-3 mb-2 fw-light"><span class="fw-bold me-2">Demo</span> Coding Spatiotemporal GNNs.</h5>
-                <p>Overview of open-source Pytorch libraries for graph-based spatiotemporal data processing and short
-                    demo
-                    with Torch Spatiotemporal.</p>
             </div>
         </div>
     </div>
@@ -156,22 +139,20 @@ <h1>Organizers</h1>
                 <div class="row g-4 my-4 justify-content-center align-items-center text-center text-md-start">
                 </div>
                 <div class="row g-4 my-4 justify-content-center align-items-top text-center">
-                    {% assign ca = people | where: "id", "calippi" | first %}
+                    {% assign ac = people | where: "id", "acini" | first %}
                     <div class="col-10 col-sm-6 col-md-3 text-center">
-                        <img class="people-thumb medium-thumb" src="{{site.url}}/assets/img/people/{{ca.img}}" />
-                        <h4 class="mt-2">{{ca.name}} <strong>{{ca.surname}}</strong></h4>
-                        <p class="fw-light text-muted mb-1">Professor</p>
-                        {% if ca.links %}
-                        {% include links_list.html links=ca.links %}
+                        <img class="people-thumb medium-thumb" src="{{site.url}}/assets/img/people/{{ac.img}}" />
+                        <h4 class="mt-2">{{ac.name}} <strong>{{ac.surname}}</strong></h4>
+                        <p class="fw-light text-muted mb-1">Post-doc Researcher</p>
+                        {% if ac.links %}
+                        {% include links_list.html links=ac.links %}
                         {% endif %}
                     </div>
 
-                    {% assign dz = people | where: "id", "dzambon" | first %}
-                    {% include people_item.html person=dz hide_description=true col_md=3 %}
-                    {% assign ac = people | where: "id", "acini" | first %}
-                    {% include people_item.html person=ac hide_description=true col_md=3 %}
                     {% assign im = people | where: "id", "imarisca" | first %}
                     {% include people_item.html person=im hide_description=true col_md=3 %}
+                    {% assign dz = people | where: "id", "dzambon" | first %}
+                    {% include people_item.html person=dz hide_description=true col_md=3 %}
                 </div>
             </div>
         </div>