Seminar Artificial Intelligence

The seminar is aimed at students interested in science in the M.Sc. Study, as well as at Ph.D. Computer engineering and engineering, computer engineering and systems, internet technologies, smart systems. The aim of the seminar is to discuss the scientific work within the scope of a doctoral program. The basis of the lectures and elaborations are therefore concrete scientific questions. The thesis includes the preparation of the relevant literature for a specific topic as well as the scientific processing of these research questions.


Prof. Dr. techn. Wolfgang Nejdl
Professorinnen und Professoren


In order to ensure intensive support during the seminar, the number of participants in the seminar is limited. The participants should have relevant lectures in the field to be treated. Ongoing work (as part of a master's thesis, diploma thesis or dissertation) can and should be incorporated into the work within the seminar.

Topics in the current semester include

  • Search engine and ranking procedures
  • Semantic Web and metadata
  • Information retrieval algorithms
  • Data mining
  • Social web
  • Human Computation and Crowdsourcing

Other topics can be included by appointment.

The seminar has two SWS for the fields of study

  • computer science
  • Mathematics / computer science
  • Electrical Engineering / Technical Computer Science

eligible, as well as additional deepening and supervision in the context of a dissertation study in the mentioned topic ranges meaningfully.

We meet in the L3S two to three days a week. If you would like to join the research seminar, please send a short email to Wolfgang Nejdl.

The list of papers that we will be discussing this semester

  • (KDD) HOLMES: Health OnLine Model Ensemble Serving for Deep Learning Models in Intensive Care Units
  • (KDD) Learning to Cluster Documents into Workspaces Using Large Scale Activity Logs
  • (KDD) DETERRENT: Knowledge Guided Graph Attention Network for Detecting Healthcare Misinformation
  • (AAMAS) Employing Models of Human Social Motor Behavior for Artificial Agent Trainers
  • (KDD) Shop The Look: Building a Large Scale Visual Shopping System at Pinterest
  • (AAMAS) Trajectory-User Linking with Attentive Recurrent Network
  • (KDD) Interleaved Sequence RNNs for Fraud Detection
  • (KDD) Aligning Superhuman AI with Human Behavior: Chess as a Model System
  • (AAMAS) HMMs for Anomaly Detection in Autonomous Robots
  • (KDD) TinyGNN: Learning Efficient Graph Neural Networks
  • (KDD) An Automatic Approach for Generating Rich, Linked Geo-Metadata from Historical Map Images
  • (KDD) Attentional Multi-graph Convolutional Network for Regional Economy Prediction with Open Migration Data


Current information on the specific dates will be published via the mailing list.

Note: The students that want to be added to the mailing list of the course, please send an e-mail to Erick Elejalde stating your interest. We want to use this list as an additional method to notify any relevant information.