A5040830236- Artificial Intelligence in Games
- Data Visualization and Analytics
- Music Technology and Sound Studies
- Creativity in Education and Neuroscience
- Scientific Computing and Data Management
- Music and Audio Processing
- Aesthetic Perception and Analysis
- Computer Graphics and Visualization Techniques
- Generative Adversarial Networks and Image Synthesis
- Neuroscience and Music Perception
- Educational Games and Gamification
- Model-Driven Software Engineering Techniques
- Computability, Logic, AI Algorithms
- Teaching and Learning Programming
- Innovative Human-Technology Interaction
- AI-based Problem Solving and Planning
- Design Education and Practice
- Neural dynamics and brain function
- Semantic Web and Ontologies
- Software Engineering Research
- Advanced Text Analysis Techniques
- Neural Networks and Reservoir Computing
- Advanced Memory and Neural Computing
- Neural Networks and Applications
- Constraint Satisfaction and Optimization
Queen Mary University of London
2019-2024
Monash University
2019-2021
Imperial College London
2008-2021
Falmouth University
2018-2021
Australian Regenerative Medicine Institute
2021
Goldsmiths University of London
2009-2018
University of Dundee
2014
University of Edinburgh
2000-2002
We describe ANGELINA3, a system that can automatically develop games along defined theme, by selecting appropriate multimedia content from variety of sources and incorporating it into game's design. discuss these capabilities in the context FACE model for assessing progress building creative systems, how ANGELINA3 be improved through further work.
Transformer models have made great strides in generating symbolically represented music with local coherence. However, controlling the development of motifs a structured way global form remains an open research area. One reasons for this challenge is due to note-by-note autoregressive generation such models, which lack ability correct themselves after deviations from motif. In addition, their structural performance on datasets shorter durations has not been studied literature. study, we...
Deep learning has enabled remarkable advances in style transfer across various domains, offering new possibilities for creative content generation. However, the realm of symbolic music, generating controllable and expressive performance-level transfers complete musical works remains challenging due to limited datasets, especially genres such as jazz, lack unified models that can handle multiple music generation tasks. This paper presents ImprovNet, a transformer-based architecture generates...
The term serendipity describes a creative process that develops, in context, with the active participation of agent, but not entirely within agent's control. While system cannot be made to perform serendipitously on demand, we argue its $\mathit{serendipity\ potential}$ can increased by means suitable architecture and other design choices. We distil unified description serendipitous occurrences from historical theorisations creativity. This takes form framework six phases:...
Text-to-image generative models have recently exploded in popularity and accessibility. Yet so far, use of these creative tasks that bridge the 2D digital world creation physical artefacts has been understudied. We conduct a pilot study to investigate if how text-to-image can be used assist upstream within process, such as ideation visualization, prior sculpture-making activity. Thirty participants selected materials generated three images using Stable Diffusion generator, each with text...
Explainable AI has the potential to support more interactive and fluid co-creative systems which can creatively collaborate with people. To do this, creative models need be amenable debugging by offering eXplainable (XAI) features are inspectable, understandable, modifiable. However, currently there is very little XAI for arts. In this work, we demonstrate how a latent variable model music generation made explainable; specifically extend MeasureVAE generates measures of music. We increase...