Michael Spranger

A key feature of intelligence in humans is that we are able to accomplish and learn wide ranging behaviors and tasks -- from moving in the world to mathematics and science. Until we have machines that are as versatile, open-ended learners as humans, we have not understood fundamental aspects of human nature. In my research, I am trying to solving the elementary puzzles that surround open-ended, incremental learning and adaptation.
One of my focus areas is language. How is language acquired by learners and how does it change on short and long timescales? I explore issues like these using autonomous robots that develop and evolve aspects of human language. My work leads to artificial systems remarkably flexible and resilient to noise, while at the same time able to achieve complex tasks. Results of this research are applied in artificial assistants, Biomedical Natural Language Processing and gaming. I also build robot sculptures and installations to start a discussion on scientific and societal issues raised by recent developments in Artificial Intelligence.

Artificial Intelligence / Machine Learning / Developmental Robotics / Computational Linguistics / Agent-based Models of the Emergence of Symbol Systems / Neural-Symbolic Systems / Reinforcement Learning





Selected Publications

Spranger, M. (2012). Potential stages in the cultural evolution of spatial language. In Scott-Phillips, T. C., Tamariz, M., Cartmill, E. A., and Hurford, J., editors, The Evolution of Language: Proceedings of the 9th International Conference (EVOLANG9).

Spranger, M., Pauw, S., Loetzsch, M., and Steels, L. (2012). Open-ended Procedural Semantics. In Steels, L. and Hild, M., editors, Language Grounding in Robots, pages 153-172. Springer.

Steels, L. and Spranger, M. (2008). The robot in the mirror. Connection Science, 20(4):337-358.


Michael Spranger received his Diploma from the Humboldt-Universität zu Berlin (Germany) in 2008 and a PhD from the Vrije Universiteit in Brussels (Belgium) in 2011 (both in Computer Science). For his PhD, Michael was a researcher at Sony CSL Paris. His PhD was honorably mentioned in 2011 among the best AI dissertations that year by the European Coordinating Committee for Artificial Intelligence. He then worked in the R&D department of Sony corporation in Tokyo (Japan) for 2 years. Currently, he is a researcher at Sony Computer Science Laboratories Inc. Michael is a roboticist by training with extensive experience in research on and construction of autonomous systems including research on robot perception, world modeling and behavior control. After his diploma he fell in love with the study of language and has since worked on models and theories of language processing, development and evolution for various language domains including action language, posture verbs, tense and aspect, determination and spatial language. Michael is a pioneer in the fields of computational construction grammar and computational cognitive semantics and a main developer of some of the most advanced computational formalisms in these fields - Fluid Construction Grammar and Incremental Recruitment Language.