The Self-Organizing Map is a vital part of today's research towards unsupervised learning in Neural Networking. There are many commercial and scholarly applications for Self-Organizing Maps within data analysis, topological visualization, and clustering. In this paper we address the functionality of the Self-Organizing Map in its tendency towards greediness. It is found that the common solution...

Specifying a numeric reward function for reinforcement learning typically requires a lot of hand-tuning from a human expert. In contrast, preference-based reinforcement learning (PBRL) utilizes only pairwise comparisons between trajectories as a feedback signal, which are often more intuitive to specify. Currently available approaches to PBRL for control problems with continuous state/action sp...

2008 Wiley Periodicals, Inc. Complexity 7: 7–13, 2008 E volutionary biology has traditionally made much of the concept of selfishness. Charles Darwin reduced adaptation (and to a less satisfying degree, speciation) to the heritable variation and competition among individuals. As understanding of the genetic control of heredity and morphogenesis has become more sophisticated, Richard Dawkins has...

This paper shows how to formally characterize language learning in a finite parameter space as a Markov structure, hnportant new language learning results follow directly: explicitly calculated sample complexity learning times under different input distribution assumptions (including CHILDES database language input) and learning regimes. We also briefly describe a new way to formally model (rap...

This paper describes the development rationale and the architecture of a prototypical expert-database system. Knowledge processing capabilities of SQL were enhanced by extending the language by recursive views. This work is based on an evolutionary approach; smooth integration with the base language was an important development aim. After a discussion of the main design alternatives, the archit...