Schedule | Speaker | Title | Slides |
---|---|---|---|
9:00 9:30 |
Welcome - Breakfast | ||
9:30 9:45 |
Organizers | Introduction | Download |
9:45 10:45 |
Fatemeh Salehi Rizi Universität Passau |
Exploiting Graph Embeddings for Graph Analysis Tasks | Download |
10:45 11:00 |
Coffee break | ||
11:00 12:00 |
Aleksandar Bojchevski Technical University of Munich |
Uncertainty and Robustness of Graph Embeddings | Download |
12:00 12:30 |
Rémi Vaudaine Université de Lyon (LHC) |
What network properties are really captured by graph embeddings? | Download |
12:30 13:30 |
Lunch | ||
13:30 14:30 |
Anton Tsitsulin Hasso Plattner Institut, Universität Potsdam |
Similarities and representations of graphs | Download |
14:30 15:00 |
Sébastien Lerique Université de Lyon (LIP/INRIA) |
Joint graph-feature embeddings using GCAEs | Download |
15:00 15:30 |
Sandra Mitrovic (Visio) KU Leuven |
Graph Embeddings in Practice: A Telco Churn Prediction Use Case | Download |
15:30 16:00 |
Coffee break | ||
16:00 16:30 |
Robin Brochier Université de Lyon (ERIC) |
Global Vectors for Text-Enhanced Networks | Download |
16:30 17:30 |
Benjamin Piwowarski CNRS (LIP6/Sorbonne Universités) |
Gaussian Embeddings for Relational Data | Download |
17:30 18:00 |
Rémy Cazabet Université de Lyon (LIRIS) |
On link prection and graph embedding | Download |
Graph Embedding is a recently introduced technique whose goal is to transform graph data into vector data
by keeping some aspects of the graph structure.
Because it allows to apply all the range of data mining and machine learning techniques
-that require vectors as input- to graph data,
it has the potential to be a game changer in the related fields of graph mining/network analysis.
Because the embedding of such a complex structure is a hard problem, it also represents a challenge
in the field of machine learning itself, in line with recent works on word embedding (word2vec...)
The aim of this event is to have some of the most active international researchers in the field to present their recent works, their vision of the field and of its future.