Szczegóły publikacji
Opis bibliograficzny
Universal intelligence, creativity, and trust in emerging global expert systems / Andrzej M. J. SKULIMOWSKI // W: Artificial Intelligence and Soft Computing : 12th International Conference, ICAISC 2013 : Zakopane, Poland, June 9–13, 2013 : proceedings, Pt. 2 / eds. Leszek Rutkowski [et al.]. — Berlin ; Heidelberg : Springer-Verlag, cop. 2013. — (Lecture Notes in Computer Science ; ISSN 0302-9743 ; 7895. Lecture Notes in Artificial Intelligence). — ISBN: 978-3-642-38609-1; e-ISBN: 978-3-642-38610-7. — S. 582–592. — Bibliogr. s. 591–592, Abstr. — Dodatkowa afiliacja autora: International Centre for Decision Sciences and Forecasting, Progress & Business Foundation, Kraków
Autor
Słowa kluczowe
Dane bibliometryczne
ID BaDAP | 74405 |
---|---|
Data dodania do BaDAP | 2013-07-16 |
DOI | 10.1007/978-3-642-38610-7_53 |
Rok publikacji | 2013 |
Typ publikacji | materiały konferencyjne (aut.) |
Otwarty dostęp | |
Konferencja | 12th International Conference on Artificial Intelligence and Soft Computing |
Czasopismo/seria | Lecture Notes in Computer Science |
Abstract
This paper presents a hypothesis together with evidence related to the use of global knowledge as a holistic expert system. By global expert system (GES) we mean all knowledge sources, bases, repositories, and processing units, regardless of whether they are human, artificial, animal, or hybrid, such that the relation "able to transfer knowledge on immediate demand" forms a connected graph over the elements of the system. A key requirement is that problem solving using GES is an anytime process with respect to the number of information sources taken into account. We conjecture that due to the high and ever-growing level of interconnection of knowledge units, a universal intelligence emerges, which under specific conditions can outperform the intelligence and creativity of any of its individual elements, including humans. It will be shown that this is possible only if an appropriate level of credibility can be assigned to each element of the system, which ensures that users trust the responses. We will design a hybrid supervised-reinforced learning scheme that makes it possible to achieve a satisfactory level of trust in GES query responses. Query processing will apply knowledge fusion methods such as combinations of recommendations and forecasts.