Constructivist learning is a hierarchical part-to-whole learning system used by humans and desirable for use by robots. Current implementations are too resource-intensive to be used for anything but simple environments. In this paper, we reimplement one such system, the Schema Mechanism, using a hierarchy of self-organizing maps. The result is an efficient system for learning perceptual and action schemas that can be used in real-world applications like robotics.

PDF

In Proceedings of WSOM '03: Workshop for Self-Organizing Maps, Kitakyushu, Japan, 2003.

@inproceedings{chaput:wsom03, title={Constructivist Learning: A Neural Implementation of the Schema Mechanism}, author={Harold H. Chaput and Benjamin Kuipers and Risto Miikkulainen}, booktitle={Proceedings of WSOM '03: Workshop for Self-Organizing Maps}, address={Kitakyushu, Japan}, url="http://www.cs.utexas.edu/users/ai-lab/?chaput:wsom03", year={2003} }