< KI2021 – Program


44th German Conference on Artificial Intelligence, September 27 – Oct 1, 2021: Online Conference

Schedule of the KI2021 program.

Please refer to your order notification from INFORMATIK-Team and click on the order link provided with the email. Then press the button "Join online event" to log into the Venueless conference system. In Venueless you find room names in the bar on the left. Links to room names in Venueless are also indicated below in the schedule. Just click on the room names in Venueless to join respective sessions (but you need to log into Venueless first).

Further presentations and social events are shown in the INFORMATIK 21 Schedule.

Monday (September 27, 2021)
9:30-10:00 Grand Opening Informatik 2021, KI2021, enviroInfo 2021 (INFORMATIK: main stage)
10:00-11:30 W3
(Room Alan Newell)
(Room David Hubel)
Early Career Research Consortium
(Room Torsten Wiesel)
12:00-13:30 W3
(Room Alan Newell)
(Room David Hubel)
14:30-16:00 T2
(Room David Hubel)
16:30-18:00 W4 (16:30-18:30)
(Room John McCarthy)
(Room David Hubel)
Wednesday (September 29, 2021)
9:00-11:00 Informatik 2021 Welcome and first session (INFORMATIK: main stage)
11:15-12:00 Kristian Kersting: The Third Wave of AI (Joint Keynote INFORMATIK 2021 and KI2021, INFORMATIK: main stage)
12:00-12:45 Artificial Intelligence, Computer Science, and Sustainability Panel Discussion (INFORMATIK: main stage)
13:45-14:00 Welcome (Room John McCarthy)

Agents and Humans 1

(Room John McCarthy, Session chair: Ralf Möller)
  • Explanation as a Process: User-centric Construction of Multi-level and Multi-modal Explanations Bettina Finzel, David Elias Tafler, Stephan Scheele, and Ute Schmid
  • Demystifying Artificial Intelligence for End-Users: Findings from a Participatory Machine Learning Show Katharina Weitz, Ruben Schlagowski, and Elisabeth André
  • An Agent Architecture for Knowledge Discovery and Evolution. Tezira Wanyana and Deshendran Moodley
15:30-16:30 Stuart Russell: Human-Compatible Artificial Intelligence (Keynote, Room John McCarthy)

Agents and Humans 2

(Room John McCarthy, Session chair: Ralf Möller)
  • RP-DQN: An Application of Q-Learning to Vehicle Routing Problems Ahmad Bdeir, Simon Boeder, Tim Dernedde, Kirill Tkachuk, Jonas K. Falkner, and Lars Schmidt-Thieme
  • Quantified Boolean Solving for Achievement Games Steve Boucher and Roger Villemaire
Thursday (September 30, 2021)

General AI 1

(Room John McCarthy, Session chair: Ralf Möller)
  • Recent Advances in Counting and Sampling Markov Equivalent DAGs Marcel Wienöbst, Max Bannach, and Maciej Liskiewicz
  • k-Circulant Maximum Variance Bases Christopher M.A. Bonenberger, Wolfgang Ertel, and Markus Schneider

General AI 2

(Room John McCarthy, Session chair: Stefan Edelkamp)
  • An Approach to Reduce the Number of Conditional Independence Tests in the PC Algorithm Marcel Wienöbst and Maciej Liskiewicz
  • Negation in Cognitive Reasoning Claudia Schon, Sophie Siebert, and Frieder Stolzenburg
  • EVARS-GPR: EVent-triggered Augmented Refitting of Gaussian Process Regression for Seasonal Data Florian Haselbeck and Dominik G. Grimm
  • Selective Pseudo-label Clustering Louis Mahon and Thomas Lukasiewicz
  • Knowledge Graph Based Question Answering System for Financial Securities Marius Bulla, Lars Hillebrand, Max Lübbering, and Rafet Sifa
  • Semantic Segmentation of Aerial Images using Binary Space Partitioning Daniel Gritzner and Jörn Ostermann
13:45-14:45 Katja Mombaur: Motion Intelligence for Human-centred Robots (Keynote, Room John McCarthy)
15:15-open end

Poster Session

(self-hosted, go to Room John McCarthy for the link, Session chair: Elmar Rückert)
  • Unsupervised Anomaly Detection for Financial Auditing with Model-Agnostic Explanations Sebastian Bruckert and Günter Pesch
  • A Priori Approximation of Symmetries in Probabilistic Dynamic Relational Models Nils Finke and Marisa Mohr
  • Towards More Explicit Interaction Modelling in Agent-based Simulation Using Affordance Schemata Franziska Klügl and Sabine Timpf
  • Critic Guided Segmentation of Rewarding Objects in First-Person Views Andrew Melnik, Augustin Harter, Christian Limberg, Krishan Rana, Niko Sünderhauf, and Helge Ritter
  • Self-supervised Domain Adaptation for Diabetic Retinopathy Classification with Constrained Loss Functions Duy Nguyen Ho Minh, Truong Thanh Nhat Mai, Ngoc Trong Tuong Than, Alexander Prange, and Daniel Sonntag
  • BehavE: Behaviour Understanding through Automated Generation of Situation Models Teodor Stoev and Kristina Yordanova
Friday (October 1, 2021)
9:00-10:00 Birte Glimm: Ontologies for Providing Map Knowledge to Autonomous Vehicles (Keynote, Room John McCarthy)

Neural Networks and Deep Learning

(Room John McCarthy, Session chair: Elmar Rückert)
  • Semi-unsupervised Learning: An In-depth Parameter Analysis Padraig Davidson, Florian Buckermann, Michael Steininger, Anna Krause, and Andreas Hotho
  • A high-speed Neural Architecture Search Considering the Number of Weights Yamada Fuyuka, Tsuji Satoki, Kawaguchi Hiroshi, Inoue Atsuki, and Sakai Yasufumi
  • Combining Transformer Generators with Convolutional Discriminators Ricard Durall, Stanislav Frolov, Andreas Dengel, and Janis Keuper
  • Learning to Detect Adversarial Examples Based on Class Scores Tobias Uelwer, Felix Michels, and Oliver De Candido
  • Interactive Image Clustering and Fine-tuning Neural Networks in VR Alexander Prange and Daniel Sonntag
  • HUI-Audio-Corpus-German: A high quality TTS dataset Pascal Puchtler, Johannes Wirth, and René Peinl
13:45-14:45 Tristan Cazenave: Monte Carlo Search (Keynote, Room John McCarthy)

Image Processing

(Room John McCarthy, Session chair: Stefan Edelkamp)
  • Crop it, But Not Too Much: The Effects of Mask-based Image Preprocessing on the Classification of Skin Lesions Fabrizio Nunnari, Abraham Obinwannne, and Daniel Sonntag
  • Multi-Type-TD-TSR - Extracting Tables from Document Images Using a Multi-stage Pipeline for Table Detection and Table Structure Recognition: From OCR to Structured Table Representations Pascal Fischer, Alen Smajic, Alexander Mehler, and Giuseppe Abrami
16:00-17:00 Giuseppe De Giacomo: Autonomy in AI: Reactive Synthesis, Planning and Reinforcement Learning in Linear Temporal Logic on Finite Traces (Keynote, Room John McCarthy)
17:00-17:30 Closing (Room John McCarthy)