Discovery of abstract concepts by a robot

Authors:
Ivan Bratko
Affiliations:
University of Ljubljana, Faculty of Computer and Information Sc., Ljubljana, Slovenia
Venue:
DS'10 Proceedings of the 13th international conference on Discovery science
Year:
2010

Citing 3
Cited 0

Prolog (3rd ed.): programming for artificial intelligence

Prolog (3rd ed.): programming for artificial intelligence
Markov logic networks

Machine Learning
An Experiment in Robot Discovery with ILP

ILP '08 Proceedings of the 18th international conference on Inductive Logic Programming

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper reviews experiments with an approach to discovery through robot's experimentation in its environment. In addition to discovering laws that enable predictions, we are particularly interested in the mechanisms that enable the discovery of abstract concepts that are not explicitly observable in the measured data, such as the notions of a tool or stability. The approach is based on the use of Inductive Logic Programming. Examples of actually discovered abstract concepts in the experiments include the concepts of a movable object, an obstacle and a tool.