SM 445/CS 707: Data and Web Science Seminar (HWS 2021)

The Data and Web Science seminar covers recent topics in data and web science. This term's topic is graph learning with a special (but not exclusive) focus on graph neural networks.

Organization

  • This seminar is organized by Prof. Dr. Rainer Gemulla and Adrian Kochsiek.
  • Available for Master students (2 SWS, 4 ECTS) and Bachelor students (2 SWS, 5 ECTS).
  • Prerequisites for Master students: solid background in machine learning
  • Maximum number of participants is 5 BSc and 10 MSc students

Goals

In this seminar, you will

  • Read, understand, and explore scientific literature
  • Summarize a current research topic in a concise report (10 single-column pages + references)
  • Give two presentations about your topic (3 minutes flash presentation, 15 minutes final presentation)
  • Moderate a scientific discussion about the topic of one of your fellow students
  • Review a (draft of a) report of a fellow student

Schedule

  • Register as described below.
  • Attend the online kickoff meeting on Sep 14, 17:15 (tentative).
  • Work individually throughout the semester according to the seminar schedule.
  • Meet your advisor for guidance and feedback.

Registration

Register via Portal 2 until Sep 6.

If you are accepted into the seminar, provide at least 4 topics of your preference (your own and/or example topics; see below) by Sep 12 via email to Adrian Kochsiek. The actual topic assignment takes place soon afterwards; we will notify you via email. Our goal is to assign one of your preferred topics to you.

Topics

Each student works on a topic within the area of the seminar along with an accompanying reference paper. Your presentation and report should explore the topic with an emphasis on the reference paper, but not just the reference paper.

We provide example topics and reference papers below. If you want, you may suggest a different reference paper (let us know after the topic assignment) or a different topic within the graph learning area (talk to us before the topic assignments). A good starting point is recent research papers in top data mining and machine learning conferences (e.g., try NeurIPS, ICLR, ICML, KDD).

Introductory lecture videos on graph learning (part of IE 674) will be made available to all participants. The following review articles may serve as further starting points:

Topic list

All topics marked as “BSc topics” can only be taken by BSc students. Unmarked topics are selected for MSc students, but can also be taken by BSc students with the appropriate background.

  1. Graph centrality (BSc topic)
    Zaki and Meira Jr.
    Ch. 4 of Data Mining and Analysis: Fundamental Concepts and Algorithms
    Cambridge University Press, March 2020
  2. Graph models / generators (BSc topic)
    Zaki and Meira Jr.
    Ch. 4 of Data Mining and Analysis: Fundamental Concepts and Algorithms
    Cambridge University Press, March 2020
  3. Graph pattern mining (BSc topic)
    Zaki and Meira Jr.
    Ch. 11 of Data Mining and Analysis: Fundamental Concepts and Algorithms
    Cambridge University Press, March 2020
  4. Spectral clustering (BSc topic)
    Zaki and Meira Jr.
    Ch. 16 of Data Mining and Analysis: Fundamental Concepts and Algorithms
    Cambridge University Press, March 2020o
  5. Graph learning benchmarks (BSc topic)
    Hu et al.
    Open Graph Benchmark: Datasets for Machine Learning on Graphs
    2020
  6. Message passing in graph neural networks
    Hamilton et al.
    Inductive Representation Learning on Large Graphs
    NIPS 2017
  7. Spectral graph convolutional networks
    Defferrard et al.
    Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering
    NIPS 2016
  8. Scaling graph neural networks
    Zeng et al.
    GraphSAINT: Graph Sampling Based Inductive Learning Method
    ICLR 2020
  9. Graph kernels and graph similarity
    Ok
    A Graph Similarity for Deep Learning
    NeurIPS 2020
  10. Expressivity of graph neural networks
    Xu et al.
    How Powerful Are Graph Neural Networks?
    ICLR 2019
  11. Graph pooling
    Mesquita et al.
    Rethinking pooling in graph neural networks
    NeurIPS 2020
  12. Pretraining graph neural nets
    Hu et al.
    Strategies for Pre-Training Graph Neural Networks
    ICLR 2020
  13. Generative models for graphs
    You et al.
    GraphRNN: Generating Realistic Graphs with Deep Auto-regressive Models
    ICML 2018
  14. Knowledge graph embeddings
    Sun et al.
    RotatE: Knowledge Graph Embedding By Relational Rotation In Complex Space
    ICLR 2019
  15. Scene graph generation
    Tang et al.
    Unbiased Scene Graph Generation From Biased Training
    CVPR 2020
  16. Graph to sequence learning
    Xu et al.
    Graph2Seq: Graph to Sequence Learning with Attention-based Neural Networks
    2018
  17. Multi-relational graphs
    Vashishth at al.
    Composition-based Multi-Relational Graph Convolutional Networks
    ICRL 2020

Supplementary materials and references