Lehr- und Forschungseinheit für Datenbanksysteme

Breadcrumb Navigation


Master Seminar "Deep Learning for Graphs" / "Recent Developments in Data Science" (WS 2019/20)




  • Contact: Prof. Dr. Thomas Seidl, Max Berrendorf, Evgeniy Faerman
  • Required: Successful participation in the lectures "Machine Learning" or "Deep Learning and Artificial Intelligence" or equivalent
    Please indicate in the central registration form in uni2work in which semesters you had attended these courses.
  • Audience: The lecture is directed towards highly motivated Master students in Mediainformatics, Bioinformatics and Informatics as well as Data Science with strong interest in cutting edge research
  • Registration: via uni2work (from Aug 26)


The availability of large amounts of data and increased computational power enabled the renaissance of neural networks as an universal paradigm for machine learning. While initially mainly boosted by advances on image classification, Deep Learning methods recently progress into different domains. One domain of particular interest are graphs, as they allow encoding less regular structures prevalent in numerous application domains, reaching from sensor networks to knowledge graphs.

In this seminar, students will discuss fundamental works in the quickly evolving area of deep learning for graphs, and in particular knowledge graphs, and will venture into the state-of-the-art of this exciting research area.


Date Time Location Description
We, 16.10.2019 14:15-15:45 131 (Oe 67) Kickoff Slides
We, 11.12.2019 14:15-15:45 131 (Oe 67) Group Presentations I
  • Knowledge Graph Embeddings
  • Homophily-based Embeddings on Non-attributed Graphs
We, 18.12.2019 14:15-15:45 131 (Oe 67) Group Presentations II
  • Graph Neural Networks
  • Entity Alignment / Graph Alignment
  • Recommender Systems
We, 15.01.2020 14:15-15:45 131 (Oe 67)

Individual Presentations I

  1. Structural-RNN: Deep learning on spatio-temporal graphs
    A. Jain, A. R. Zamir, S. Savarese, A. Saxena
  2. Graph Representation Learning via Hard and Channel-Wise Attention Networks
    Hongyang Gao, Shuiwang Ji
  3. Dynamic Spatial-Temporal Graph Convolutional Neural Networks for Traffic Forecasting
    Zulong Diao, Xin Wang, Dafang Zhang, Yingru Liu, Kun Xie, Shaoyao He
  4. How powerful are graph neural networks?
    Keyulu Xu, Weihua Hu, Jure Leskovec, Stefanie Jegelka
  5. Is a Single Embedding Enough? Learning Node Representations that Capture Multiple Social Contexts.
    Alessandro Epasto, Bryan Perozzi
We, 22.01.2020 14:15-15:45 131 (Oe 67)

Individual Presentations II

  1. Iterative Deep Graph Learning for Graph Neural Networks
    Yu Chen, Lingfei Wu, Mohammed J. Zaki
    ICLR'20 (reject)
  2. Uncertainty-Aware Prediction for Graph Neural Networks
    Xujiang Zhao, Feng Chen, Shu Hu, jin-Hee Cho
    ICLR'20 (reject)
  3. NetGAN: Generating Graphs via Random Walks
    Aleksandar Bojchevski, Oleksandr Shchur, Daniel Zügner, Stephan Günnemann
  4. Position-aware Graph Neural Networks
    Jiaxuan You, Rex Ying, Jure Leskovec
  5. Embedding Logical Queries on Knowledge Graphs
    William L. Hamilton, Payal Bajaj, Marinka Zitnik, Dan Jurafsky, Jure Leskovec
We, 29.01.2020 14:15-15:45 131 (Oe 67)

Individual Presentations III

  1. Entity Alignment between Knowledge Graphs Using Attribute Embeddings
    Bayu Distiawan Trisedy, Jianzhong Qi, Rui Zhang
  2. Predicting Dynamic Embedding Trajectory in Temporal Interaction Networks
    Srijan Kumar, Xikun Zhang, Jure Leskovec
  3. Graph U-Nets
    Hongyang Gao, Shuiwang Ji
  4. Simplifying Graph Convolutional Networks
    Felix Wu, Tianyi Zhang, Amauri Holanda de Souza Jr., Christopher Fifty, Tao Yu, Kilian Q. Weinberger
  5. GNNExplainer: Generating Explanations for Graph Neural Networks
    Rex Ying, Dylan Bourgeois, Jiaxuan You, Marinka Zitnik, Jure Leskovec
Su, 01.03.2020 Seminar Thesis Submission Deadline

Additional Information: