Artificial K-Lines and Applications - Advances in Computational Science and Engineering - page 97

space can be utilized to automate the input of a human expert. In addition to the game of
Tic-Tac-To, it is interesting to formulate the automation of human expert input for more
complex games, such as chess and go. Another interesting research direction as an ap-
plication of AKL, is the area of artificial creativity . Note, an inherent characteristic of
AKL is to combine parts of K-lines and form new K-lines comprised of those parts. We
argue that this is the essence of creativity, that is, the ability to form new “ideas” by
using existing knowledge, or old ideas. In the context of the AKL, the newly formed K-
lines represent the new ideas, whereas the existing knowledge is the sequence of seg-
ments of the existing K-lines that are used to form the new K-lines. In this sense, the
AKL might be a suitable candidate structure for artificial creativity. It is our intention to
delve into this line of research. One major obvious issue is how to judge the suitability
of the newly generated K-lines, i.e., how to zero-in into only the meaningful generated
ideas among all generated ideas, since the latter can be too many and, among those,
many/most of them might not make sense. Note, again, the same phenomenon is en-
countered with natural creativity. That is, for every new idea that is generated by a per-
son, there is always the issue of how sensible (or applicable) that idea is. We will be
delighted if the AKL can shed some light to the artificial creativity problem, which is a
many-decades old and unsolved problem in Artificial Intelligence.
References
1. Minsky, M.: K-Lines: A Theory of Memory. Cognitive Science 4, 117-133 (1980)
2. Minsky, M.: The Society of Mind. Simon & Schuster (1986)
3. Minsky, M.: The Emotion Machine. Simon & Schuster (2006)
4. Sloman, A.: Grand Challenge 5: The Architecture of Brain and Mind: Integrating Low-
Level Neuronal Brain Processes with High-Level Cognitive Behaviours in a Functioning
Robot. Technical Report COSY-TR-0607 (July 2006),
http://www.cs.bham.ac.uk/research/projects/cosy/papers/
#tr0607
5. Singh, P.: Failure Directed Reformulation, M. Eng. Thesis, Massachusetts Institute of Tech-
nology, Department of Electrical Engineering and Computer Science (March 10, 1998)
6. Singh, P.: EM-ONE: An Architecture for Reflective Commonsense Thinking, Ph.D. Thesis,
Massachusetts Institute of Technology, Department of Electrical Engineering and Computer
Science (June 2005)
7. Toptsis, A.A., Dubitski, A.: Iterative K-line Meshing via Non-Linear Least Squares Interpo-
lation of Affectively Decorated Media Repositories. The Open Artificial Intelligence Jour-
nal 2, 46-61 (2008)