[May 27] Invitation talk dr Rob Saunders, Motivated, Social and Embodied Computational Creativity
Posted: 22 May 2019 11:24 AM
Administrator
RankRankRankRank
Total Posts:  175
Joined  2011-09-08

Herewith you are cordially invited for attending the following lecture by Dr. Rob Saunders.

Date and time: 27 May 10:00
Location: Snellius room 312

Motivated, Social and Embodied Computational Creativity

Abstract:
This lecture will provide an overview of my research across Computational Creativity, Creative Computing and Creative Robotics. In particular, I will present my work developing computational models of creativity that are intrinsically motivated, socially situated and physically embodied.

My work in Computational Creativity has explored the role of intrinsic motivation, e.g., curiosity, in computational models of creativity. My research developing computational models of curiosity, using a variety of techniques from machine learning, has opened up new avenues for research in computational creativity, supporting the development of “curious design agents” and novel modes of interaction with “curious assistants”. Results from this research suggest that intrinsically motivated agents can effectively explore novel design spaces, and support conversational interactions with designers. Multi-agent simulations with curious design agents support new computational models of social and cultural creativity that display phenomena observed in creative societies.

My work in Creative Computing has explored the development of online learning platforms for creative coding that encourage the playful use of code as a medium of expression alongside formal learning material. Results from this research suggest that facilitating informal learning in the classroom strengthens the building of productive learning networks that benefit the whole cohort. More recently, my research has explored the development of parametric design tools for novice game designers, incorporating “inspirational” computational creativity systems that support novel forms of social learning, e.g., “rapid game jams”, where the process of designing and developing a game is reduced from hours to minutes permitting rapid iteration of ideas and the emergence of novel meta-game design interactions between game jam participants.

My work in Creative Robotics has explored the development of computational creativity as embodied systems. My work in this area has included the creation of robotic artworks that perform the curious exploration of materially engaged learning on an architectural scale. My on-going research in Performative Body Mapping (PBM) uses computational models of curiosity for grounded demonstration learning in non-anthropomorphic social robots through collaboration with movement experts from choreography and dance. The PBM methodology permits a social robot to learn the constraints for expressive movement from experts while maintaining the ability for improvisation. Results from studies with experts and non-experts demonstrate that PBM facilitates rich and engaging human-robot interactions with abstract, non-humanlike machines.

Profile