XAI
MT4H International workshop
The ‘Machine Teaching for Humans’ workshop took place at Universidade de Madeira, during the days of 11th to 15th of January 2023. The workshop featured talks from keynote speakers Jerry Zhu, Patrick Shafto, Ute Schmid, and Sebastian Lapuschkin, as well as other invited speakers. A panel discussion on MT4H was held. Around 30 participants attended the workshop, including project members from UiB, VRAIN, Equinor and Eviny.
Featured talks
The workshop was pleased to welcome the following speakers.
- Keynote speakers:
- Jerry Zhu (University of Wisconsin-Madison) – Machine Teaching and “Nature Vs. Nurture”
- Patrick Shafto (Rutgers University) – Explanable AI as cooperative communication: Theory and data
- Sebastian Lapuschkin (Fraunhofer HHI) – Human-understandable explanations through Concept Relevance Propagation
- Ute Schmid (University of Bamberg) – Near-miss explanations to teach humans and machines
- Invited talks:
- Adish Singla (Max Planck Institute for Software Systems) – Adaptive Scaffolding in Block-Based Programming
- Eduardo Fermé (University of Madeira) – Dynamics of profiles from a KR perspective
- Hans Simon (Max Planck Institute Saarbrücken and Ruhr-Uni-Bochum) – Collusion-free Machine Teaching
- Peter Flach (University of Bristol) – XAI for time series
- Raul Santos-Rodriguez (University of Bristol) – Humans, labels and explanations
- Téo Sanchez (Université Paris Saclay) – Humans as teachers: An HCI approach to building ML models
- Shared talks:
- César Ferri (Technical University of Valencia) – Teaching size
- Jakob Vigerust Kallestad (Equinor) – Evaluation of Text Classification tasks with Equinor’s maintenance reports
- Jan Arne Telle (University of Bergen) – Probabilistic Machine Teaching
- Kristian Flikka (Eviny AS) – A matter of trust
- Manuel G. Piqueras (University of Castilla – La Mancha) – Heuristic search in Machine Teaching
Full list of participants
- Adish Singla (Max Planck Institute for Software Systems)
- Anna-Lena Both (Eviny AS)
- Bjarte Johansen (Equinor)
- Brigt Arve T. Håvardstun (University of Bergen)
- Cèsar Ferri (Technical University of Valencia)
- Christian Askvik-hansen (Eviny AS)
- Diogo Nuno Freitas (University of Madeira)
- Eduardo Fermé (University of Madeira)
- Gloria Palencia (Technical University of Valencia)
- Hans Simon (Max Planck Institute Saarbrücken and Ruhr-Uni-Bochum)
- Henrik Olsen Eide (University of Bergen)
- Jakob Vigerust Kallestad (Equinor)
- Jan Arne Telle (University of Bergen)
- Jerry Zhu (University of Wisconsin-Madison)
- Jorge Fernandes (University of Madeira)
- José Hernández-Orallo (Technical University of Valencia / University of Cambridge)
- Kristian Flikka (Eviny AS)
- Magrete Torland (Equinor)
- Manuel G. Piqueras (University of Castilla – La Mancha)
- Pål Grønås Drange (University of Bergen)
- Patrick Shafto (Rutgers University)
- Pekka Parviainen (University of Bergen)
- Peter Flach (University of Bristol)
- Raul Santos-Rodriguez (University of Bristol)
- Sebastian Lapuschkin (Fraunhofer HHI)
- Seong-Eun Cho (Equinor)
- Téo Sanchez (Université Paris Saclay)
- Ute Schmid (University of Bamberg)
“Machine Teaching for Explainable AI” is a joint project between the Department of Informatics at the University of Bergen, the Valencian Research Institute for AI (VRAIN) and industrial partners Equinor and Eviny. The project is financed by the Norwegian Research Council.