Evaluating machine creativity

PLOS ONE, 2016

Creativity is a complex, multi-faceted concept encompassing a variety of related aspects, abilities, properties and behaviours. If we wish to study creativity scientifically, then a tractable and well-articulated model of creativity is required. Such a model would be of great value to researchers investigating the nature of creativity and in particular, those concerned with the evaluation of creative practice. This paper describes a unique approach to developing a suitable model of how creative behaviour emerges that is based on the words people use to describe the concept. Using techniques from the field of statistical natural language processing, we identify a collection of fourteen key components of creativity through an analysis of a corpus of academic papers on the topic. Words are identified which appear significantly often in connection with discussions of the concept. Using a measure of lexical similarity to help cluster these words, a number of distinct themes emerge, which collectively contribute to a comprehensive and multi-perspective model of creativity. The components provide an ontology of creativity: a set of building blocks which can be used to model creative practice in a variety of domains. The components have been employed in two case studies to evaluate the creativity of computational systems and have proven useful in articulating achievements of this work and directions for further research. 'The philosophy, science and engineering of computational systems which, by taking on particular responsibilities, exhibit behaviours that unbiased observers would deem to be creative.' [1, p. 21] Computational creativity research follows both theoretical and practical directions and crosses several disciplinary boundaries between the arts, sciences, and engineering.

Connection Science, 2016

Free download during September 2016 from http://tandfonline.com/doi/full/10.1080/09540091.2016.1151860 Computational creativity is the modelling, simulating or replicating of creativity computationally. In examining and learning from these “creative systems”, from what perspective should the creativity of a system be considered? Are we interested in the creativity of the system's output? Or of its creative processes? Features of the system? Or how it operates within its environment? Traditionally computational creativity has focused more on creative systems' products or processes, though this focus has widened recently. Creativity research offers the Four Ps of creativity: Person/Producer, Product, Process and Press/Environment. This paper presents the Four Ps, explaining each in the context of creativity research and how it relates to computational creativity. To illustrate the usefulness of the Four Ps in taking broader perspectives on creativity in its computational treatment, the concepts of novelty and value are explored using the Four Ps, highlighting aspects of novelty and value that may otherwise be overlooked. Analysis of recent research in computational creativity finds that although each of the Four Ps appears in the body of computational creativity work, individual pieces of work often do not acknowledge all Four Ps, missing opportunities to widen their work's relevance. We can see, though, that high-status computational creativity papers do typically address all Four Ps. This paper argues that the broader views of creativity afforded by the Four Ps is vital in guiding us towards more comprehensively useful computational investigations of creativity. Available at http://www.tandfonline.com/doi/full/10.1080/09540091.2016.1151860

Proceedings of the 7th Computational Creativity Symposium at AISB 2021, 2021

This philosophical paper examines the Darwinian account of creativity as a model for assessing computational creativity. It will first establish a Darwinian account of creativity using Simonton’s model. It will then apply this model to popular image-producing AI, Generative Adversarial Networks, and the promising Creative Adversarial Network, both used in the computational production of ‘artworks’. The paper will argue that these networks are compatible with a Darwinian account of creativity, due to the presence of blind variation within the networks, a key component of Simonton’s model. The paper will then address some initial objections. The aim of this paper will ultimately be to assess whether the AI systems are compatible with the Darwinian model of creativity, and in the process explore Darwinian creativity as a potential standard for testing computational creativity.

Proceedings of xCoAx2020, 2020

The question of whether a machine can be creative has been at the center of many scholarly debates. But what does the public think about the possibility for AI to gain a place alongside human artists? This paper presents the results of a survey conducted at the University of Nottingham which investigated the public reception of the application of Artificial Intelligence to the creative sector. The study examined the attitudes and beliefs of participants to the prospect of a future scenario where machines create art alongside and in collaboration with humans. The responses, collected both through an online questionnaire and a focus group, reveal that participants do not exclude the possibility that in the medium-term AI may earn the attribute ‘creative’. Still, this does not mean that this scenario is welcomed.

2002

Over the last few years, the number of large volume databases has increased dramatically. Consequently, it is much more difficult to understand the retrieved often multivariate information. User-adaptive visualization systems could assist this process. This paper presents data from a pilot study of user preferences for different types of information visualizations under experimental conditions where they were allowed to choose between alternative informationequivalent representations in order to answer given questions. The results are being used to inform the design of a user modeling component of AIVE, an adaptive system for information display.

2019

Generative design methods are gaining importance in all the design industries. Responding to a detected lack of research on generative design (GD), especially in graphic design, this thesis provides an overview of key concepts, history, areas of application and state of the research on the topic. Based on insights from this overview, four essential unanswered questions are addressed, leading to some main conclusions about the present and future of generative design. Generative design is found to be a revolutionary method of creation that initiates a new era in design and possibly humankind, making it very probable for it to last in time. As a main reason for graphic design lagging behind in the application of generative processes, the concept of the “linguistic gap” is introduced. There is a great linguistic difference between the psychological and aesthetical concepts on which graphic design problems are based and the algorithmic language of computer code. This makes it difficult to define and evaluate generative processes in graphic design. Concerning the changing role of human designers in the generative design process, algorithmic thinking and programming are identified as essential skills for designers in the near future. Addressing a further away future, it is stated that a complete substitution of human designers by generative Artificial Intelligence (AI) systems is probable. In a critical debate, the implied economic, social, cultural and ethical problems in this kind of future are discussed. As the main approach to prevent possible problematic consequences of generative design in the future, a change in design education and general education is recommended. Lastly, a practical application of generative processes in graphic design is proposed in the layout of the thesis document and conclusions are drawn from the corresponding learning and creation process.

2002

ABSTRACT This paper describes the development of a design methodology for informing the design of new technologies for the domestic environment that harness the potential affordances of broadband technology. With reference to one possible new broadband service,'SnapShots', a prototype application is being developed. This is a household visualisation and planning tool for major family events and will be evaluated in a home setting.

Foreword from the Congress Chairs, 2012

Abstract. The field of behaviour-based artificial intelligence (AI), with its roots in the robotics research of Rodney Brooks, is not predominantly tied to linguistic interaction in the sense of the classic Turing test (or,“imitation game”). Yet, it is worth noting, both are centred on a behavioural model of intelligence. Similarly, there is no intrinsic connection between musical AI and the language-based Turing test, though there have been many attempts to forge connections between them. Nonetheless, there are aspects of musical AI and the Turing ...

International Journal of Technology Enhanced Learning, 2014

in 2012. Since 2010 he works as Research Assistant at the Department of Psychology and at the University of Graz and is involved in national and international scientific projects. He is currently involved in the EU GALA Network of Excellence (www.galanoe.eu). His research topics are neuronal plasticity through learning, EEG-based neurofeedback and auditory mirror neurons. Furthermore Manuel Ninaus has many years of experience with different neurophysiological methods such as EEG, NIRS and fMRI.

Keyphrases. Age. Gender. Intelligence Types. Turing Test Taxonomy. Theory of Multiple Intelligences. Meta-modular intelligence. Autonomous Artificial Agent. Attribution of civil rights to informatic systems. Chintamani fractal model of intelligence type&component clustering. BTTT and ETTT techspiecies annotation schemas 5