The First Annual Contact Forum of the PROSECCO network will be held in El Escorial, Madrid from February 21-23 inclusive. The forum will bring together international researchers whose work can inform CC research and be informed in turn by ongoing CC efforts.
Computational Design and Visualization Lab (PDF)
Penousal Machado, University of Coimbra, Portugal
This presentation will make a short overview of ongoing research at the CDV Lab. of the University of Coimbra, including topics such as: computational art and design, computational aesthetics, natureinspired computing, information visualization.
EMRG /research between art, design & science (PDF)
Lucas Nijs, Sint Lucas School of Arts, Antwerp, Belgium
EMRG research activities include individual PhD dissertations funded by the St Lucas University College of Art and Design and the University of Antwerp, cutting-edge art & science projects with competively acquired funding, and shorter projects involving external partners and companies.
A Computational Model of Analogical Reasoning in Dementia Care (PDF)
Konstantinos Zachos, City University, London, UK
A practical application of a computational model of analogical reasoning to a pressing social problem, which is to improve the care of older people with dementia. Underpinning the support for carers for people with dementia is a computational model of analogical reasoning that retrieves information about cases from analogical problem domains. The model implements structure-mapping theory adapted to match source and target domains expressed in unstructured natural language. The model is implemented as a computational service invoked by a mobile app used by carers during their care shifts.
Emergence, synchonicity and generative processes in visual and sound creativity (PDF)
Jaime Munarriz, Universidad Complutense de Madrid
Generative audio-visual systems based on:
• A-life : autonomous agents, morphology (form/function), evolution, connections
• Node networks : pulses, microsequencers, synch, evolving morphology
• Emergence (desired)
• Generative language: collage/remix, tweeter derive, Markov chains
• Associative content: text / sound / images. Direct - conflicting – evocative. Evolutive systems as an associative content creation mechanism for live performance.
Analogy, Concept Blending, and Computational Creativity (PDF)
Tarek Besold, University of Osnabruck, Germany
Analogy and concept blending are often considered crucial parts of concept invention and creative processes in general. In this presentation I will shortly outline the basics of the Heuristic-Driven Theory Projection (HDTP) computational analogy framework, show how the approach can be expanded to also perform concept blending, and provide a rough sketch of the computational cognitive model envisioned to serve as core for the COINVENT project.
Computational Creativity through Conceptual Integration (PDF)
Mark Turner, Case Western University, USA
I will sketch an approach to computational creativity through conceptual integration, also called "blending," as outlined in Turner, Mark (2014) , The Origin of Ideas: Blending, Creativity, and the Human Spark (Oxford University Press).
John Barnden, University of Birmingham, UK
I will comment on various types of creativity in metaphor, and mention how the ATT-Meta approach to metaphor understanding handles some types. One key feature of the approach is the downplaying of the importance of creating new analogical mappings. Another key aspect is the handling of metaphor compounds of various different sorts. I like the idea that metaphor is used within the mind, not just in communication, and therefore could/should be within AI systems' minds too. Metaphor-in-mind may contribute to creativity more broadly than implied by theories that implicate analogy creation in creativity.
Creative Natural Language Processing (PDF)
Carlo Strapparava, University of Trento, Italy
Dealing with creative language and in particular with affective, persuasive and even humorous language has often been considered outside the scope of computational linguistics. Nonetheless it is possible to exploit current NLP techniques starting some explorations about it. We briefly review some computational experiences about these typical creative genres.
Computational Stylometry (PDF)
Walter Daelemans, University of Antwerp, Belgium
I will give a brief overview of work at CLiPS (University of Antwerp) on Computational Stylometry: detecting psychological and sociological properties of the author of a text (age, gender, personality, education level, psychological health etc.) on the basis of linguistic properties of the text. I will also discuss some possible applications in Computational Creativity.
Universality and Creativity in Language: Author profiling and Irony (PDF)
Paolo Rosso, Universitat Politécnica de Valencia, Spain
I will give a brief overview of work at NLE Lab (PRHLT research center of the Technical University of Valencia), especially focussing on what we did in irony detection in social media and on what we started recently doing wrt author profiling in social media in order to discuss how the use of language, including figurative devices such as irony, varies among classes of authors and depends on authors’ demographics such as age and gender.
Story generation and co-creation with the Virtual Storyteller (PDF)
Mariët Theune, University of Twente, The Netherlands
The Virtual Storyteller is an emergent narrative system, in which stories emerge from the interactions of autonomous intelligent agents in a simulated story world. The actions of the agents are captured in the form of a causal network, from which story texts can be generated. Human players can take over control of the characters, enabling a form of co-creation with the system. We have performed experiments with children interacting with the system.
Interfacing Imagination (PDF)
Licinio Roque, University of Coimbra, Portugal
This presentation will give a quick overview of our ongoing research in game design and media design, including topics such as game design, sound design, author-centric gameplay generation, procedural content generation in games, and personalized game-based learning, the goal of participatory media and how it relates to creativity and the idea of interfacing imagination.
Applying Computational Creativity Approaches to Developmental Robotics (PDF)
Georgi Stojanov, American University of Paris, France
In Developmental Robotics the main goal of the researchers is to find a good learning algorithm which will allow an anthropomorphic physical robot (usually highly complex system with high-dimensional sensory-motor space) to come up with a feasible high-level descriptions of their environment. These algorithms tend to be heavily biased towards looking for structures in the sensory-motor flux, very expensive computationally, because of the high dimensionality of sensory-motor space and lacking significant collection of innate structures and/or biases.
In this context, our proposal is to reverse the current approaches in Developmental Robotics, and just like in Computational Creativity, provide the agent with a rich library of biases and behaviors (preferably not computationally expensive) which the agent will use to impose a structure on the sensory-motor flow in a weakly supervised way. Thus, instead of only looking for structures in a bottom up level (starting with the raw sensory-motor data) and essentially be at mercy of the outside world, the agent will be co-creating its various ontologies. We call this process projection and it can be related to the Piagetian notion of assimilation. There is a huge body of research from Developmental Psychology which suggests the plausibility of this approach. Ideally, the agent will have the possibility to switch from one ontology to another, depending on its goals and, more broadly, its internal state.
Computational Temperature: the most underexploited mechanism of computational creativity (PDF)
Robert French, CNRS/University of Burgundy, France
Temperature almost certainly takes the cake for being both an incredibly important mechanism for computational creativity, yet at the same time, an incredibly overlooked one. We will discuss computational temperature, why it is important to computational creativity, and how it can be implemented easily in programs involving any kind of decision making. A brief example or two will be given of how to put this mechanism to use.
Mathematics and creativity: an empirical approach (PDF)
Ursula Martin, Queen Mary University of London, UK
For centuries, the highest level of mathematics has been seen as an isolated creative activity, to produce a proof for review and acceptance by research peers. Mathematics is now at a remarkable inflexion point, with new technology radically extending the power and limits of individuals. “Crowdsourcing” pulls together diverse experts to solve problems; symbolic computation tackles huge routine calculations; and computers, using programs designed to verify hardware, check proofs that are just too long and complicated for any human to comprehend.
Yet these techniques are currently used in stand-alone fashion, lacking integration with each other or with human creativity or fallibility. “Social machines” are new paradigm, identified by Berners-Lee, for viewing a combination of people and computers as a single problem-solving entity.
Is it possible to measure the aesthetics of photographs? (PDF)
Stefan Rüger, Open University, UK
This presentation looks at methods to recognise one aspect of human creativity, the algorithmic assessment of visual aesthetics of photographs. Once this can be done to a suitable degree of satisfaction, the computational creation of visual photographs or visual artwork, is a big step closer, for example through genetic algorithms that use the computational aesthetics model as fitness function. We will briefly look at methods to derive suitable high and low-level features from photographic theory and from machine learning that allow to assess the photographic quality of an image in terms of simplicity, realism and compositional principles.
Computational Creativity in Games (PDF1, PDF2)
Julian Togelius, ITU Copenhagen, Denmark
Georgios N. Yannakakis, University of Malta, Malta
I discuss the opportunities for computational creativity in computer games, and in particular link thinking about computational creativity to the now substantial volume of ongoing work on procedural content generation in games. Creativity is meaningful in the context of audio, graphics, levels, game rules and gameplay itself, but even more so in the context of complete games which are multifaceted artefacts of considerable complexity and artistic value. Moreover, I will argue that games is one of the few domains where there is a clear commercial case for computational creativity. I will start giving a general introduction and Georgios talks about some more specific project.
The ER Model: from plot generation to visual composition (PDF)
Rafael Perez y Perez, Universidad Autónoma Metropolitana, Mexico
My work on CC has mainly focused on plot generation based on the ER Model. In the last years I have also started to develop computer models for visual composition. In this presentation I’ll summarize 15 years of research on CC and I will describe our current efforts in the production and evaluation of narratives and visual compositions.
Experimenting Human Creativity by means of Unplugged Evolutionary Algorithms (PDF)
Francisco Fernandez Vega, University of Extremadura, Spain
It is not easy to find examples of artistic collaboration aiming at generating an artwork that smoothly integrates individual influences from all of the artists. This experience, that embodies Evolutionary Algorithm (EA) rules, has allowed a team of artists to generate a collective work; Yet, each of the individual artwork is the result of individual artist characteristic creative experience that helps the team to enhanced the creative process: an iterative process that includes concepts such as crossover and mutation, has helped artist to express their creativity in a collective framework. This specific unplugged and heterodox version of an Evolutionary Algorithm provides new means for artist inspiration while allowing researchers more deeply studying the underlying principles of human creativity from an evolutionary perspective.
Therefore, this artistic experience shows how artists inspiration can benefit from an EA based methodology while artist's creative methods, knowledge and experience expressed along the work, may help in the future to develop better EAs devoted to art and design.
The potential of memorized experiences for creative problem solving and learning (PDF)
Agnar Aamodt, Norwegian University of Science and Technology, Norway
Memory-based, or case-based, reasoning makes use of past, concrete experiences in understanding present situations and solving current problems. Several case-based approaches to creativity have been suggested, including the tweaking of a specific experience or pattern, and the combining of casespecific and generalized knowledge in novel ways. The talk will briefly summarize parts of this research and suggest some important questions to guide further investigation.
Computational creativity and surprise in data mining / Music informatics and computational creativity (PDF)
Tijl De Bie, University Of Bristol, UK
In the first half of this presentation I briefly will talk about my work on formalising interestingness of data mining patterns by quantifying the amount of surprise this pattern elicits in the data miner. I will speculate about possible connections between creativity and surprise, and thus how this work could inform computational creativity research. In the second half, I will talk about my work on music informatics, in particular on past work on quantifying the potential of a song to become a chart hit, and on ongoing work to create an artificial DJ.