Skip to content

Commit 252e448

Browse files
FilippoMBFilippoMB
committed
fixed thesis page
1 parent 3a0c9f3 commit 252e448

File tree

34 files changed

+1533
-223
lines changed

34 files changed

+1533
-223
lines changed

‎_config.yml‎

Lines changed: 4 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -59,10 +59,12 @@ collections:
5959
news:
6060
output: true
6161
permalink: /news/:path/
62+
theses:
63+
output: true
64+
permalink: /theses/:path/
6265

6366
# Set to `true` to show excerpts on the homepage.
64-
#
6567
show_excerpts: true
6668

6769
# display posts with dates in the future
68-
future: true
70+
future: true

‎_data/people.yml‎

Lines changed: 18 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -31,6 +31,7 @@ phd:
3131
scholar: https://scholar.google.ca/citations?user=HzMTPo0AAAAJ&hl=en&oi=ao
3232
linkedin: https://www.linkedin.com/in/michele-guerra-34a273182/
3333
github: https://github.com/MicheleUIT
34+
webpage: "https://en.uit.no/ansatte/person?p_document_id=767125&p_dimension_id=88140"
3435

3536
- name: "Jonas Berg Hansen"
3637
bio: "Jonas is a PhD student researching pooling in graph neural networks. He applies his research in the context of energy analytics."
@@ -59,3 +60,20 @@ collaborator:
5960
github: "https://github.com/spindro"
6061
linkedin: "https://www.linkedin.com/in/indro-spinelli-770380123/"
6162
twitter: "https://x.com/IndroSpinelli"
63+
64+
- name: "Veronica Lachi"
65+
image: "/figs/people/vl.png"
66+
position: "Postdoc"
67+
affiliation: "Fondazione Bruno Kessler"
68+
scholar: "https://scholar.google.com/citations?user=uly8D-sAAAAJ&hl=it"
69+
linkedin: "https://www.linkedin.com/in/veronica-lachi-85b726254/"
70+
twitter: "https://x.com/lachiveronica"
71+
github: "https://github.com/veronicalachi"
72+
73+
- name: "Carlo Abate"
74+
image: "/figs/people/ca.jpeg"
75+
position: "PhD Student"
76+
affiliation: "Università di Bologna, IIT Genova"
77+
scholar: "https://scholar.google.com/citations?user=s6gZ2LAAAAAJ&hl=it"
78+
github: "https://github.com/carloabate"
79+
linkedin: "https://www.linkedin.com/in/carloabate/"

‎_events/2024-06-05-talk-giacomo-boracchi.md‎

Lines changed: 40 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,40 @@
1+
---
2+
title: "QuanTrees: Histograms for Monitoring Multivariate Data Streams"
3+
layout: post
4+
date: 2024-06-05 10:15
5+
place: "Forskningsparken: Klasserom 1 (Forskningsparken) B281"
6+
presenter: "Giacomo Boracchi"
7+
affiliation: "Politecnico di Milano, Italy"
8+
published: true
9+
---
10+
11+
# QuanTrees: Histograms for Monitoring Multivariate Data Streams
12+
13+
## Abstract
14+
Change detection problems are ubiquitous in science and engineering: promptly detecting changes is often crucial to understanding the dynamics of a monitored process and activating suitable countermeasures. During this talk, I will present QuantTree (QT), a particular type of histogram yielding a very practical and powerful change detection algorithm for monitoring multivariate data streams. The significant advantages of QuantTree are its nonparametric nature, which allows adapting to any stationary distribution, and its theoretical properties, which enable setting a priori the detection thresholds without resorting to bootstrap. I will then illustrate QT-EWMA, an efficient extension of QT that combines QT with an Exponential Weighted Moving Average scheme to perform sequential monitoring. I will conclude my talk with Class Distribution Monitoring (CDM), an effective concept-drift detection scheme that uses QT-EWMA to monitor the class-conditional distributions of a data stream.
15+
16+
## Presenter
17+
- Name: Giacomo Boracchi
18+
- Affiliation: Politecnico di Milano, DEIB - Dipartimento di Elettronica, Informazione e Bioingegneria - Milano, Italy
19+
- Webpage: https://boracchi.faculty.polimi.it/
20+
21+
### Bio
22+
Giacomo Boracchi is Associate Professor of Computer Engineering at Dipartimento di Elettronica, Informazione e Bioingegneria of the Politecnico di Milano (DEIB). Giacomo has received a Ph.D. in information technology (DEIB, 2008), and an MSc degree in Mathematics (Universitá Statale di Milano, 2004). His primary research interests concern image processing and machine learning, in particular, image restoration and analysis, change/anomaly detection, domain adaptation, and learning in nonstationary environments. He has been/currently is the advisor of 8 Ph.D. students, regularly teaching deep learning and image processing courses and giving tutorials at major conferences (ICASSP, ICIP, ICPR, IJCNN).
23+
24+
He is currently leading research projects applying machine learning and image-understanding algorithms to design new industrial solutions with STMicroelectronics, Gilardoni Raggi X, and Cisco. He is the author of more than 90 papers in international conferences and journals, and since 2018 he has served as an Associate Editor for the IEEE Transactions on Image Processing. In 2015 he received the IBM Faculty Award; in 2016 the IEEE Transactions on Neural Networks and Learning Systems Outstanding Paper Award; in 2017, the Nokia Visiting Professor Scholarship; in 2021, an nVidia Applied Research Grant.
25+
26+
27+
## References
28+
- Giacomo Boracchi, Diego Carrera, Cristiano Cervellera, Danilo Macciò "QuantTree: Histograms for Change Detection in Multivariate Data Streams" International Conference Machine Learning (ICML) 2018
29+
- [paper](http://proceedings.mlr.press/v80/boracchi18a.html)
30+
- [code](https://github.com/diegocarrera89/quantTree)
31+
32+
- Luca Frittoli, Diego Carrera and Giacomo Boracchi, "Nonparametric and Online Change Detection in Multivariate Datastreams using QuantTree" IEEE Transactions on Data Knowledge and Engineering, 2022,
33+
- [paper](https://boracchi.faculty.polimi.it/docs/2022_08_TKDE_Sequential_QTree_Frittoli.pdf)
34+
- [code](https://github.com/diegocarrera89/quantTree)
35+
- [project page](https://boracchi.faculty.polimi.it/Projects/QT-EWMA.html)
36+
37+
- Diego Stucchi, Luca Frittoli, Giacomo Boracchi "Class Distribution Monitoring for Concept Drift Detection" IEEE-INNS International Joint Conference on Neural Networks (IJCNN) 2022
38+
- [preprint](https://boracchi.faculty.polimi.it/docs/2022_01_IJCNN_Class_Distribution_Monitoring_CDM.pdf)
39+
- [code](https://github.com/diegocarrera89/quantTree)
40+
- [project page](https://boracchi.faculty.polimi.it/Projects/CDM_Class_Distribution_Mo)

‎_news/preprint_koopman_stgnn.md‎

Lines changed: 16 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,16 @@
1+
---
2+
title: "New preprint"
3+
layout: post
4+
date: 2024-10-15
5+
published: true
6+
---
7+
8+
"*Interpreting Temporal Graph Neural Networks with Koopman Theory*"
9+
10+
By Michele Guerra, Simone Scardapane, and Filippo Maria Bianchi.
11+
12+
<!--more-->
13+
14+
The paper proposes a XAI technique based on Koopman theory to interpret temporal graphs and spatio-temporal GNNs. The proposed approach allows to identify nodes and time steps when relevant events occur.
15+
16+
The preprint is available on [Arxiv](https://arxiv.org/abs/2410.13469) and the code on [Github](https://github.com/NGMLGroup/Koopman-TGNN-Interpretability).

‎_news/preprint_maxcut.md‎

Lines changed: 16 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,16 @@
1+
---
2+
title: "New preprint"
3+
layout: post
4+
date: 2024-09-10
5+
published: true
6+
---
7+
8+
"*MaxCutPool: differentiable feature-aware Maxcut for pooling in graph neural networks*"
9+
10+
By Carlo Abate and Filippo Maria Bianchi.
11+
12+
<!--more-->
13+
14+
The paper proposes a novel approach to compute a MaxCut partition in attributed graphs and can be used to implement pooling in Graph Neural Networks. It works particulary well on heterophilic graphs.
15+
16+
The preprint is available on [Arxiv](https://arxiv.org/abs/2409.05100) and the code on [Github](https://github.com/NGMLGroup/MaxCutPool).

‎_site/events/2024-06-05-talk-giacomo-boracchi/index.html‎

Lines changed: 135 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,135 @@
1+
<!DOCTYPE html>
2+
<html lang="en">
3+
4+
<head>
5+
<!-- Other head content -->
6+
<script src="https://code.jquery.com/jquery-3.6.0.min.js"></script>
7+
<link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/all.min.css">
8+
</head><head>
9+
<meta charset="utf-8">
10+
<meta http-equiv="X-UA-Compatible" content="IE=edge">
11+
<meta name="viewport" content="width=device-width, initial-scale=1"><!-- Begin Jekyll SEO tag v2.8.0 -->
12+
<title>QuanTrees: Histograms for Monitoring Multivariate Data Streams | Northernmost GraphML Group</title>
13+
<meta name="generator" content="Jekyll v4.3.4" />
14+
<meta property="og:title" content="QuanTrees: Histograms for Monitoring Multivariate Data Streams" />
15+
<meta property="og:locale" content="en_US" />
16+
<meta name="description" content="QuanTrees: Histograms for Monitoring Multivariate Data Streams" />
17+
<meta property="og:description" content="QuanTrees: Histograms for Monitoring Multivariate Data Streams" />
18+
<link rel="canonical" href="http://localhost:4000/events/2024-06-05-talk-giacomo-boracchi/" />
19+
<meta property="og:url" content="http://localhost:4000/events/2024-06-05-talk-giacomo-boracchi/" />
20+
<meta property="og:site_name" content="Northernmost GraphML Group" />
21+
<meta property="og:type" content="article" />
22+
<meta property="article:published_time" content="2024-06-05T10:15:00+02:00" />
23+
<meta name="twitter:card" content="summary" />
24+
<meta property="twitter:title" content="QuanTrees: Histograms for Monitoring Multivariate Data Streams" />
25+
<script type="application/ld+json">
26+
{"@context":"https://schema.org","@type":"BlogPosting","dateModified":"2024-06-05T10:15:00+02:00","datePublished":"2024-06-05T10:15:00+02:00","description":"QuanTrees: Histograms for Monitoring Multivariate Data Streams","headline":"QuanTrees: Histograms for Monitoring Multivariate Data Streams","mainEntityOfPage":{"@type":"WebPage","@id":"http://localhost:4000/events/2024-06-05-talk-giacomo-boracchi/"},"publisher":{"@type":"Organization","logo":{"@type":"ImageObject","url":"http://localhost:4000/figs/NGMLGlogo2.png"}},"url":"http://localhost:4000/events/2024-06-05-talk-giacomo-boracchi/"}</script>
27+
<!-- End Jekyll SEO tag -->
28+
<link rel="stylesheet" href="/assets/css/style.css"><link type="application/atom+xml" rel="alternate" href="http://localhost:4000/feed.xml" title="Northernmost GraphML Group" />
29+
</head>
30+
<body><header class="site-header">
31+
32+
<div class="wrapper"><div class="branding">
33+
<a href="/" class="logo-link">
34+
<img src="/figs/NGMLGlogo2.png" alt="Northernmost GraphML Group" class="site-logo" />
35+
<span class="site-title">Northernmost GraphML Group</span>
36+
</a>
37+
</div><nav class="site-nav">
38+
<input type="checkbox" id="nav-trigger" class="nav-trigger" />
39+
<label for="nav-trigger">
40+
<span class="menu-icon">
41+
<svg viewBox="0 0 18 15" width="18px" height="15px">
42+
<path d="M18,1.484c0,0.82-0.665,1.484-1.484,1.484H1.484C0.665,2.969,0,2.304,0,1.484l0,0C0,0.665,0.665,0,1.484,0 h15.032C17.335,0,18,0.665,18,1.484L18,1.484z M18,7.516C18,8.335,17.335,9,16.516,9H1.484C0.665,9,0,8.335,0,7.516l0,0 c0-0.82,0.665-1.484,1.484-1.484h15.032C17.335,6.031,18,6.696,18,7.516L18,7.516z M18,13.516C18,14.335,17.335,15,16.516,15H1.484 C0.665,15,0,14.335,0,13.516l0,0c0-0.82,0.665-1.483,1.484-1.483h15.032C17.335,12.031,18,12.695,18,13.516L18,13.516z"/>
43+
</svg>
44+
</span>
45+
</label>
46+
47+
<div class="trigger"><a class="page-link" href="/people.html">People</a><a class="page-link" href="/publications.html">Publications</a><a class="page-link" href="/theses.html">Theses</a><a class="page-link" href="/news.html">Events</a></div>
48+
</nav></div>
49+
</header>
50+
<main class="page-content" aria-label="Content">
51+
<div class="wrapper">
52+
<article class="post h-entry" itemscope itemtype="http://schema.org/BlogPosting">
53+
54+
<header class="post-header">
55+
<h1 class="post-title p-name" itemprop="name headline">QuanTrees: Histograms for Monitoring Multivariate Data Streams</h1>
56+
<p class="post-meta"><time class="dt-published" datetime="2024-06-05T10:15:00+02:00" itemprop="datePublished">
57+
Jun 5, 2024
58+
</time></p>
59+
</header>
60+
61+
<div class="post-content e-content" itemprop="articleBody">
62+
<h1 id="quantrees-histograms-for-monitoring-multivariate-data-streams">QuanTrees: Histograms for Monitoring Multivariate Data Streams</h1>
63+
64+
<h2 id="abstract">Abstract</h2>
65+
<p>Change detection problems are ubiquitous in science and engineering: promptly detecting changes is often crucial to understanding the dynamics of a monitored process and activating suitable countermeasures. During this talk, I will present QuantTree (QT), a particular type of histogram yielding a very practical and powerful change detection algorithm for monitoring multivariate data streams. The significant advantages of QuantTree are its nonparametric nature, which allows adapting to any stationary distribution, and its theoretical properties, which enable setting a priori the detection thresholds without resorting to bootstrap. I will then illustrate QT-EWMA, an efficient extension of QT that combines QT with an Exponential Weighted Moving Average scheme to perform sequential monitoring. I will conclude my talk with Class Distribution Monitoring (CDM), an effective concept-drift detection scheme that uses QT-EWMA to monitor the class-conditional distributions of a data stream.</p>
66+
67+
<h2 id="presenter">Presenter</h2>
68+
<ul>
69+
<li>Name: Giacomo Boracchi</li>
70+
<li>Affiliation: Politecnico di Milano, DEIB - Dipartimento di Elettronica, Informazione e Bioingegneria - Milano, Italy</li>
71+
<li>Webpage: https://boracchi.faculty.polimi.it/</li>
72+
</ul>
73+
74+
<h3 id="bio">Bio</h3>
75+
<p>Giacomo Boracchi is Associate Professor of Computer Engineering at Dipartimento di Elettronica, Informazione e Bioingegneria of the Politecnico di Milano (DEIB). Giacomo has received a Ph.D. in information technology (DEIB, 2008), and an MSc degree in Mathematics (Universitá Statale di Milano, 2004). His primary research interests concern image processing and machine learning, in particular, image restoration and analysis, change/anomaly detection, domain adaptation, and learning in nonstationary environments. He has been/currently is the advisor of 8 Ph.D. students, regularly teaching deep learning and image processing courses and giving tutorials at major conferences (ICASSP, ICIP, ICPR, IJCNN).</p>
76+
77+
<p>He is currently leading research projects applying machine learning and image-understanding algorithms to design new industrial solutions with STMicroelectronics, Gilardoni Raggi X, and Cisco. He is the author of more than 90 papers in international conferences and journals, and since 2018 he has served as an Associate Editor for the IEEE Transactions on Image Processing. In 2015 he received the IBM Faculty Award; in 2016 the IEEE Transactions on Neural Networks and Learning Systems Outstanding Paper Award; in 2017, the Nokia Visiting Professor Scholarship; in 2021, an nVidia Applied Research Grant.</p>
78+
79+
<h2 id="references">References</h2>
80+
<ul>
81+
<li>Giacomo Boracchi, Diego Carrera, Cristiano Cervellera, Danilo Macciò “QuantTree: Histograms for Change Detection in Multivariate Data Streams” International Conference Machine Learning (ICML) 2018
82+
<ul>
83+
<li><a href="http://proceedings.mlr.press/v80/boracchi18a.html">paper</a></li>
84+
<li><a href="https://github.com/diegocarrera89/quantTree">code</a></li>
85+
</ul>
86+
</li>
87+
<li>Luca Frittoli, Diego Carrera and Giacomo Boracchi, “Nonparametric and Online Change Detection in Multivariate Datastreams using QuantTree” IEEE Transactions on Data Knowledge and Engineering, 2022,
88+
<ul>
89+
<li><a href="https://boracchi.faculty.polimi.it/docs/2022_08_TKDE_Sequential_QTree_Frittoli.pdf">paper</a></li>
90+
<li><a href="https://github.com/diegocarrera89/quantTree">code</a></li>
91+
<li><a href="https://boracchi.faculty.polimi.it/Projects/QT-EWMA.html">project page</a></li>
92+
</ul>
93+
</li>
94+
<li>Diego Stucchi, Luca Frittoli, Giacomo Boracchi “Class Distribution Monitoring for Concept Drift Detection” IEEE-INNS International Joint Conference on Neural Networks (IJCNN) 2022
95+
<ul>
96+
<li><a href="https://boracchi.faculty.polimi.it/docs/2022_01_IJCNN_Class_Distribution_Monitoring_CDM.pdf">preprint</a></li>
97+
<li><a href="https://github.com/diegocarrera89/quantTree">code</a></li>
98+
<li><a href="https://boracchi.faculty.polimi.it/Projects/CDM_Class_Distribution_Mo">project page</a></li>
99+
</ul>
100+
</li>
101+
</ul>
102+
103+
</div><a class="u-url" href="/events/2024-06-05-talk-giacomo-boracchi/" hidden></a>
104+
</article>
105+
106+
</div>
107+
</main><footer class="site-footer h-card">
108+
<data class="u-url" href="/"></data>
109+
110+
<div class="wrapper">
111+
112+
<div class="footer-col-wrapper">
113+
<div class="footer-col">
114+
<p class="social-links"><ul class="social-media-list"><li>
115+
<a rel="me" href="https://github.com/ngmlgroup" target="_blank" title="github">
116+
<svg class="svg-icon grey">
117+
<use xlink:href="/assets/minima-social-icons.svg#github"></use>
118+
</svg>
119+
</a>
120+
</li>
121+
</ul>
122+
</p>
123+
</div>
124+
<div class="footer-col">
125+
<p>The Northermost GraphML group in the world, based in Tromsø, Norway.
126+
</p>
127+
</div>
128+
</div>
129+
130+
</div>
131+
132+
</footer>
133+
</body>
134+
135+
</html>

‎_site/feed.xml‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,2 +1,2 @@
1-
<?xml version="1.0" encoding="utf-8"?><feed xmlns="http://www.w3.org/2005/Atom" ><generator uri="https://jekyllrb.com/" version="4.3.4">Jekyll</generator><link href="http://localhost:4000/feed.xml" rel="self" type="application/atom+xml" /><link href="http://localhost:4000/" rel="alternate" type="text/html" /><updated>2024-10-17T22:07:37+02:00</updated><id>http://localhost:4000/feed.xml</id><title type="html">Northernmost GraphML Group</title><subtitle>The Northermost GraphML group in the world, based in Tromsø, Norway.
1+
<?xml version="1.0" encoding="utf-8"?><feed xmlns="http://www.w3.org/2005/Atom" ><generator uri="https://jekyllrb.com/" version="4.3.4">Jekyll</generator><link href="http://localhost:4000/feed.xml" rel="self" type="application/atom+xml" /><link href="http://localhost:4000/" rel="alternate" type="text/html" /><updated>2024-10-20T12:21:21+02:00</updated><id>http://localhost:4000/feed.xml</id><title type="html">Northernmost GraphML Group</title><subtitle>The Northermost GraphML group in the world, based in Tromsø, Norway.
22
</subtitle></feed>

‎_site/figs/NGMLGlogo2.png‎

1.22 KB
Loading

‎_site/figs/people/ca.jpeg‎

6.64 KB
Loading

‎_site/figs/people/vl.png‎

115 KB
Loading

0 commit comments

Comments
 (0)