Towards a Computational Theory of Epistemic Creativity

0446636 - ÚI 2016 RIV GB eng C - Konferenční příspěvek (zahraniční konf.)
Wiedermann, Jiří - van Leeuwen, J.
Towards a Computational Theory of Epistemic Creativity.
41st Annual Convention of the Society for the Study of Artificial Intelligence and the Simulation of Behaviour (AISB 2015). London: Society for the Study of Artificial Intelligence & Simulation of Behaviour, 2015, s. 235-242. ISBN 978-1-5108-0386-2.
[AISB Symposium on Computing and Philosophy: The Significance of Metaphor and Other Figurative Modes of Expression and Thought /8./. Canterbury (GB), 20.04.2015-22.04.2015]
Grant ostatní: GA ČR(CZ) GA15-04960S
Institucionální podpora: RVO:67985807
Klíčová slova: epistemology * creativity * knowledge * computation
Kód oboru RIV: IN - Informatika
https://www.cs.kent.ac.uk/events/2015/AISB2015/proceedings.html

We investigate the computational process of creativity from the viewpoint of our recent thesis stating that computation is a process of knowledge generation. Rather than considering the creativity process in its full generality, we restrict ourselves to so-called epistemic creativity which deals with the processes that create knowledge. Within this domain we mainly concentrate on elementary acts of creativity - viz. drawing analogies. In order to do so using the epistemic framework, we define analogies as certain relationships among linguistic expressions and we state what knowledge must be discovered in order to resolve a given incompletely specified analogy. We assume analogies are formed in a natural language and also require that a solution of each analogy must contain an explanation why the resulting analogy holds. Finally, the difference between noncreative and creative computational processes is discussed. Our approach differs from the majority of previous approaches in stressing the knowledge discovery aspects of computational creativity, in requiring explanations in analogy solving and, last but not least, in including theory-less domains serving as knowledge base for knowledge discovery process.
Trvalý link: http://hdl.handle.net/11104/0249718