Supportive and Antagonistic Behaviour in Distributed Computational Creativity via Coupled Empowerment Maximisation

Guckelsberger, Christian, Salge, Christophe, Saunders, Rob and Colton, Simon (2016) Supportive and Antagonistic Behaviour in Distributed Computational Creativity via Coupled Empowerment Maximisation. In: Proceedings of the Seventh International Conference on Computational Creativity, France.

Preview

Text guckelsberger_iccc2016.pdf - Published Version Download (237kB) | Preview

Abstract / Summary

There has been a strong tendency in distributed computational creativity systems to embrace embodied and situated agents for their flexible and adaptive behaviour. Intrinsically mo- tivated agents are particularly successful in this respect, be- cause they do not rely on externally specified goals, and can thus react flexibly to changes in open-ended environments. While supportive and antagonistic behaviour is omnipresent when people interact in creative tasks, existing implementa- tions cannot establish such behaviour without constraining their agents’ flexibility by means of explicitly specified in- teraction rules. More open approaches in contrast cannot guarantee that support or antagonistic behaviour ever comes about. We define the information-theoretic principle of cou- pled empowerment maximisation as an intrinsically moti- vated frame for supportive and antagonistic behaviour within which agents can interact with maximum flexibility. We pro- vide an intuition and a formalisation for an arbitrary number of agents. We then draw on several case-studies of co-creative and social creativity systems to make detailed predictions of the potential effect the underlying empowerment maximisa- tion principle might have on the behaviour of creative agents.

Item Type:	Conference or Workshop Item (Paper)
Subjects:	Computer Science, Information & General Works
Courses by Department:	The Games Academy > Digital Games
Depositing User:	Simon Colton
Date Deposited:	23 Jun 2017 13:10
Last Modified:	23 Jun 2017 13:10
URI:	https://repository.falmouth.ac.uk/id/eprint/2559

Actions

View Item (login required)