Event Resources :
Paris, France | June 27th, 2016
The PROSECCO network organised a tutorial session in the 7th International Conference on Computational Creativity, in the morning of June 27th.
A flavour of Computational Creativity (CC) research was provided by introductory lectures from Geraint Wiggins, Hannu Toivonen and Graeme Ritchie. The tutorial was open to all, but was intended in first place to researchers wanting to start exploring the field, or students starting a research program in a CC related matter.
The PROSECCO network offered 11 scholarship grants especially aimed at young researchers, of up to €1,000 each, to defray the cost of attending the tutorial. Each grant also included registration at the ICCC conference, allowing attendees to immerse themselves in Computational Creativity research for the whole week.
- Characterising Computational Creativity, Geraint A. Wiggins, Queen Mary University of London, UK - download slides
- Roles of Data Mining and Machine Learning in Computational Creativity, Hannu Toivonen, University of Helsinki, Finland - download slides
- Assessing the Performance of a Creative System, Graeme Ritchie, University of Aberdeen, UK - download slides
Characterising Computational CreativityGeraint A. Wiggins, Queen Mary University of London, UK
I introduce some fundamental notions necessary to describe creative systems, clarifying some concepts presented in and arising from Margaret Boden’s (1990) descriptive hierarchy of creativity, by beginning to formalise the ideas she proposes. The aim is to move towards a model which allows detailed comparison, and hence better understanding, of systems which exhibit behaviour which would be called ‘‘creative’’ in humans. I demonstrate some simple reasoning about creative behaviour based on the new framework, to show how it might be useful for the analysis and study of creative systems. The same mechanisms potentially allow reflection. I suggest that Boden’s descriptive framework, once elaborated in detail, is more uniform and more powerful than it first appears.
Roles of Data Mining and Machine Learning in Computational CreativityHannu Toivonen, University of Helsinki, Finland
Data mining and machine learning can be used in a number of ways to help computers learn how to be creative, such as learning to generate new artefacts or to evaluate various qualities of newly generated artefacts. In this tutorial we give an overview of the roles that data mining and machine learning have had and could have in creative systems (but we will not go into the methods).
Assessing the Performance of a Creative SystemGraeme Ritchie, University of Aberdeen, UK
When any computer program has been constructed, it is natural to ask: how well does it perform? For a potentially creative computer system, this involves various tasks, such as defining the question more precisely (which is not trivial), and devising workable methods of evaluating the performance of such a system. In making the question more precise, it is essential to consider the goals of the research activity, and how the notion of “creative” is related to other concepts which might be easier to measure directly. Evaluation studies typically rely on ratings by human judges, but this leads to various methodological matters, such as how natural the setting should be, what the judges should be asked to do, and whether comparisons should be made with human-generated items. These matters have attracted much attention within computational creativity over the past 15 years or so, and the ICCC-16 sessions will no doubt include some papers on the topic of evaluation. This talk gives an introductory overview, intended to make it easier see where individual projects and particular debates fit into the wider picture.
Organized by PROSECCO