2nd International Conference on Evolutionary and Biologically Inspired Music, Sound, Art and Design

Following the success of previous events and the importance of the field of evolutionary and biologically inspired music, sound, art and design, EvoMUSART has become an EvoStar conference with independent proceedings. Thus, EvoMUSART 2013 is the eleventh European Event and the second International Conference on Evolutionary and Biologically Inspired Music, Sound, Art and Design.

Accepted Papers

The use of biologically inspired techniques for the development of artistic systems is a recent, exciting and significant area of research. There is a growing interest in the application of these techniques in fields such as: visual art and music generation, analysis, and interpretation; sound synthesis; architecture; video; poetry; design; and other creative tasks.

The main goal of evomusart 2013 is to bring together researchers who are using biologically inspired computer techniques for artistic tasks, providing the opportunity to promote, present and discuss ongoing work in the area.

The event will be held from 3-5 April, 2013 in Vienna, Austria as part of the EvoStar event.

Publication Details

Submissions will be rigorously reviewed for scientific and artistic merit. Accepted papers will be presented orally or as posters at the event and included in the evomusart proceedings, published by Springer Verlag in a dedicated volume of the Lecture Notes in Computer Science series. The acceptance rate at evomusart 2012 was 34.9% for papers accepted for oral presentation, or 46.5% for oral and poster presentation combined.

Authors of selected papers may be invited to submit extended versions of their work to the Springer journal Genetic Programming and Evolvable Machines (GPEM).

Topics of interest

Submissions should concern the use of biologically inspired computer techniques -- e.g. Evolutionary Computation, Artificial Life, Artificial Neural Networks, Swarm Intelligence, other artificial intelligence techniques -- in the generation, analysis and interpretation of art, music, design, architecture and other artistic fields.

Topics of interest include, but are not limited to:

Important Dates

Submission deadline: 1 November 2012 Extended to 11 November 2012
Conference: 3-5 April 2013

Additional information and submission details

Submit your manuscript, at most 12 A4 pages long, in Springer LNCS format no later than November 1, 2012 to site http://myreview.csregistry.org/evomusart13

The reviewing process will be double-blind; please omit information about the authors in the submitted paper.

Programme committee

 

EvoMUSART  Abstracts

Art, Aesthetics, Evolution
Jon McCormack
This paper discusses issues in evolutionary art related to Art Theory and Aesthetics with a view to better understanding how they might contribute to both research and practice. Aesthetics is a term often used in evolutionary art, but is regularly used with conflicting or naive understandings. A selective history of evolutionary art as art is provided, with an examination of some art theories from within the field. A brief review of aesthetics as studied in philosophy and art theory follows. It is proposed that evolutionary art needs to resolve some important conflicts and be clearer about what it means by terms like ``art'' and ``aesthetics''. Finally some possibilities for how to resolve these conflicts are described.

Application of an Island Model Genetic Algorithm for a Multi-Track Music Segmentation Problem
Brigitte Rafael, Michael Affenzeller, Stefan Wagner
Genetic algorithms have been introduced to the field of media segmentation including image, video, and also music segmentation since segmentation problems usually have complex search spaces. Music segmentation can give insight into the structure of a music composition so it is an important task in music information retrieval (MIR). Past approaches have applied genetic algorithms to achieve the segmentation of a single music track. However, music compositions usually contain multiple tracks so single track segmentations might miss important global structure information. This paper focuses on the introduction of an island model genetic algorithm to achieve single track segmentations with respect to the global structure of the composition.

evoDrummer: Deriving rhythmic patterns through interactive genetic algorithms
Maximos Kaliakatsos-Papakostas, Andreas Floros, Michael Vrahatis
Drum rhythm automatic construction is an important step towards the design of systems which automatically compose music. This work describes a novel mechanism that allows a system, namely the evoDrummer, to create novel rhythms with reference to a base rhythm. The user interactively defines the amount of divergence between the base rhythm and the generated ones. The methodology followed towards this aim incorporates the utilization of Genetic Algorithms and allows the evoDrummer to provide several alternative rhythms with specific, controlled divergence from the selected base rhythm. To this end, the notion of rhythm divergence is also introduced, based on a set of 40 drum--specific features. Four population initialization schemes are discussed and an extensive experimental evaluation is provided. The obtained results demonstrate that, with proper population initialization, the evoDrummer is able to produce a great variety of rhythmic patterns which accurately encompass the desired divergence from the base rhythm.

Darwinian Pianos: Realtime Composition based on Competitive Evolutionary Process
Guido Kramann
In this project a composition is achieved by two separate evolutionary algorithms (virtual pianists) executing and modifying a repetitive phrase in a cooperative manner - conversely this collaboration is directly counteracted by deliberate placement of a tone within the repetitive phrasing by one or other of the pianists. This action creates conflict and consequently it becomes a challenging task for the opposing pianist to introduce a similar change - thus the effect becomes combative and may be witnessed by an audience. The genetic representation for pitches is based on prime-number ratios and assigns lower Hamilton distances to more harmonically related frequency pairs. This and a special way to evaluate musical structure based on it seems to be correlated with good results in generated music pieces. Finally possibilities are discussed to bring "Darwinian Pianos" into musical practice.

Finding Image Features Associated with High Aesthetic Value by Machine Learning
Vic Ciesielski, Perry Barile, Karen Trist
A major goal of evolutionary art is to get images of high aesthetic value.  We assume that some features of images are associated with high aesthetic value and want to find them.  We have taken two image databases that have been rated by humans, a photographic database and one of abstract images generated by evolutionary art software.  We have computed 55 features for each database.  We have extracted two categories of rankings, the lowest and the highest.  Using feature extraction methods from machine learning we have identified the features most associated with differences.  For the photographic images the key features are wavelet and texture features.  For the abstract images the features are colour based features.

Aesthetic Measures for Evolutionary Vase Design
Kate Reed
In order to avoid the expense of interactive evolution, some researchers have begun using aesthetic measures as fitness functions. This paper explores the potential of one of the earliest aesthetic measures by George Birkhoff as a fitness function in vase design after suitable modifications. Initial testing of vases of this form also revealed several other properties with a positive correlation with human-awarded scores. A suitable balance of these new measures along with Birkhoff's measure was found using feedback from volunteers, and vases evolved using the measure were also assessed for their aesthetic potential. Although the initial designs suffered from lack of diversity, some modifications led to a measure that enabled the evolution of a range of vases which were liked by many of the volunteers. The final range of vases included many shapes similar to those developed by human designers. Coupled with 3D printing techniques this measure allows automation of the whole process from conception to production. We hope that this demonstration of the theory will enable further work on other aesthetic products.

Inverse Mapping with Sensitivity Analysis for Partial Selection in Interactive Evolution
Jonathan Eisenmann, Matthew Lewis, Rick Parent
Evolutionary algorithms have shown themselves to be useful interactive design tools.  However, current algorithms only receive feedback about candidate fitness at the whole-candidate level.  In this paper we describe a model-free method, using sensitivity analysis, which allows designers to provide fitness feedback to the system at the component level.  Any part of a candidate can be marked by the designer as interesting (i.e. having high fitness).  This has the potential to improve the design experience in two ways: (1) The finer-grain guidance provided by partial selections facilitates more precise iteration on design ideas so the designer can maximize her energy and attention. (2) When steering the evolutionary system with more detailed feedback, the designer may discover greater feelings of satisfaction with and ownership over the final designs.

Swarmic Sketches and Attention Mechanism
Mohammad Majid al-Rifaie, John Mark Bishop
This paper introduces a novel approach deploying the mechanism of `attention' by adapting a swarm intelligence algorithm -- Stochastic Diffusion Search -- to selectively attend to detailed areas of a digital canvas.  Once the attention of the swarm is drawn to a certain line within the canvas, the capability of another swarm intelligence algorithm -- Particle Swarm Intelligence -- is used to produce a `swarmic sketch' of the attended line. The swarms move throughout the digital canvas in an attempt to satisfy their dynamic roles -- attention to areas with more details -- associated to them via their fitness function. Having associated the rendering process with the concepts of attention, the performance of the participating swarms creates a unique, non-identical sketch each time the `artist' swarms embark on interpreting the input line drawings. The detailed investigation of the `creativity' of such systems have been explored in our previous work; nonetheless, this papers provides a brief account of the `computational creativity' of the work through two prerequisites of creativity within the swarm intelligence's two infamous phases of exploration and exploitation; these phases are described herein through the attention and tracing mechanisms respectively.

Swarmic Paintings and Colour Attention
Mohammad Majid al-Rifaie, Mark Bishop
Swarm-based multi-agent systems have been deployed in non-photorealistic rendering for many years. This paper introduces a novel approach in adapting a swarm intelligence algorithm -- Stochastic Diffusion Search -- for producing non-photorealistic images. The swarm-based system is presented with a digital image and the agents move throughout the digital canvas in an attempt to satisfy the dynamic roles -- attention to different colours -- associated to them via their fitness function. Having associated the rendering process with the concepts of `attention' in general and colour attention in particular, this papers briefly discusses the `computational creativity' of the work through two prerequisites of creativity (i.e. freedom and constraints) within the swarm intelligence's two infamous phases of exploration and exploitation.

Evolving Glitch Art
Eelco den Heijer
In this paper we introduce Glitch art as a new representation in Evolutionary Art. Glitch art is a recent form of digital art, and can be considered an umbrella term for a variety of techniques that manipulate digital images by altering their digital encoding in unconventional ways.  We gathered a number of basic glitch operations and created a `glitch recipe' which takes a source image (in a certain image format, like jpeg or gif) and applies one or more glitch operations. This glitch recipe is the genotype representation in our evolutionary GP art system. We present our glitch operations, the genotype, and the genetic operators initialisation, crossover and mutation. A glitch operation may `break' an image by destroying certain data in the image encoding, and therefore we have calculated the `fatality rate' of each glitch operation. A glitch operation may also result in an image that is visually the same as its original, and therefore we also calculated the visual impact of each glitch operation. Furthermore we performed an experiment with our Glitch art genotype in our unsupervised evolutionary art system, and show that the use of our new genotype results in a new class of images in the evolutionary art world.

EvoSpace-Interactive: A Framework to Develop Distributed Collaborative-Interactive Evolutionary Algorithms for Artistic Design
Mario Garcia-Valdez, Leonardo Trujillo, Francisco Fernández de Vega, Juan Julián Merelo Guervós, Gustavo Olague
Currently, a large number of computing systems and user applications are focused on distributed and collaborative models for heterogeneous devices, exploiting cloud-based approaches and social networking. However, such systems have not been fully exploited by the evolutionary computation community. This work is an attempt to bridge this gap, and integrate interactive evolutionary computation with a distributed cloud-based approach that integrates with social networking for collaborative design of artistic artifacts. Such an approach to evolutionary art could fully leverage the concept of memes as an idea that spreads from person to person, within a computational system. In particular, this work presents EvoSpace-Interactive, an open source framework for the development of collaborative-interactive evolutionary algorithms, a computational tool that facilitates the development of interactive algorithms for artistic design. A proof of concept application is developed on EvoSpace-Interactive called Shapes that incorporates the popular social network Facebook for the collaborative evolution of artistic images generated using the Processing programming language. Initial results are encouraging, Shapes illustrates that it is possible to use EvoSpace-Interactive to effectively develop and deploy a collaborative system.

Feature Selection and Novelty in Computational Aesthetics
João Correia, Penousal Machado, Juan Romero, Adrian Carballal
An approach for exploring novelty in expression-based evolutionary art systems is presented. The framework is composed of a feature extractor, a classifier, an evolutionary engine and a supervisor. The evolutionary engine exploits shortcomings of the classifier, generating misclassified instances. These instances update the training set and the classifier is re-trained. This iterative process forces the evolutionary algorithm to explore new paths leading to the creation of novel imagery. The experiments presented and analyzed herein explore different feature selection methods and indicate the validity of the approach.

Biologically–inspired Motion Pattern Design of Multi–legged Creatures
Shihui Guo, Safa Tharib, Jian Chang, Jianjun Zhang
In this paper, we propose a novel strategy to synthesize motion patterns for multi--legged creatures inspired by the biological knowledge. To prove the concept, our framework deploys an approach of coupling the dynamics model, the Inverted Pendulum Model, and the biological controller, the Central Pattern Generator, to synthesize the motion of multiple legged creatures. The dynamics model ensures the physical plausibility and allows the virtual character to react to the external perturbations, where the biological controller coordinates the motion of several legs with designed numerical operators, providing user-friendly high--level control. This novel framework is computational--efficient by taking advantages of the self-similarity in motion and able to animate characters with different skeletons.

Decision Chain Encoding
Patrick Janssen, Vignesh Kaushik
A novel encoding technique is presented that allows constraints to be easily handled in an intuitive way. The proposed encoding technique structures the genotype-phenotype mapping process as a sequential chain of decision points, where each decision point consists of a choice between alternative options. In order to demonstrate the feasibility of the decision chain encoding technique, a case-study is presented for the evolutionary optimization of the architectural design for a large residential building.

Story Characterization Using Interactive Evolution in a Multi-Agent System
Malik Nairat, Palle Dahlstedt, Mats Nordahl
We propose a character generative approach that integrates human creativity based on an agent-based system where characters are developed using interactive evolution. By observing their behaviour, the author can choose the characters that he likes during an interaction process. The evolved characters can then be used to build a story outline as a foundation for generating stories. This can provide storytelling authors with tools for the creation process of characters and stories.

Sentient World: Human-Based Procedural Cartography
Antonios Liapis, Georgios Yannakakis, Julian Togelius
This paper presents a first step towards a computer-aided design tool for the creation of game maps. The tool, named Sentient World, allows the designer to draw a rough terrain sketch, adding extra levels of detail through stochastic and gradient search. Novelty search generates a number of dissimilar artificial neural networks that are trained to approximate a designer's sketch and provide maps of higher resolution back to the designer. As the procedurally generated maps are presented to the designer (to accept, reject, or edit) the terrain sketches are iteratively refined into complete high resolution maps which may diverge from initial designer concepts. Results obtained on a number of test maps show that novelty search is beneficial for introducing divergent content to the designer without reducing the speed of iterative map refinement.