Volker Tresp
Volker Tresp works on Bayesian machine learning and representations for relational data. He trained in physics at Göttingen and Yale, moved to Siemens in 1990 and became an LMU informatics professor in 2011. His methods include the Bayesian Committee Machine and shared work on representations such as RESCAL for multi-relational graphs. His current research covers temporal knowledge graphs, multimodal data and foundation models. Teams across LMU, Bonn and Denmark jointly developed PyKEEN.
- Now at
- Siemens and LMU Munich
- Works
- 426
- Citations
- 16,618
- h-index
- 61
Publication numbers from OpenAlex, as of 14 October 2026.
Timeline
-
1984–1990
Physics training and Siemens
Tresp earned a Göttingen physics diploma in 1984 and Yale postgraduate degrees, including a doctorate in 1989. He joined Siemens in 1990 and led machine-learning research teams.
-
1994–1997
Neural and Bayesian learning
He had a 1994 visiting-scientist role at MIT and received Siemens Inventor of the Year recognition in 1997 for neural-network research. He worked on Bayesian learning and on integrating simulated and observed industrial data.
-
2011
LMU research and teaching
He became an LMU informatics professor in 2011. He worked on the Bayesian Committee Machine and shared representation-learning methods for multi-relational graphs, including RESCAL.
-
2021
Tools for knowledge-graph learning
Teams across LMU, Bonn and Denmark collaborated on PyKEEN in 2021. His current research focus includes temporal knowledge graphs, text, images, video and foundation models. These are team research directions and tools.
Sources and credits
Photo: Slloris, via Wikimedia Commons. License: CC BY-SA 4.0. Source: commons.wikimedia.org. Cropped to a square and resized. Ask us to remove or replace this photo.
