Search Results

< Back Dr Michelangelo Conserva Queen Mary University of London iGGi Alum Michelangelo Conserva is a second year PhD researcher studying principled exploration strategies in reinforcement learning. He is particularly interested in randomized exploration and, more generally, Bayesian methods for reinforcement learning. He holds a BSc in Statistics, Economics and Finance from Sapienza, University of Rome and an MSc in Computational Statistics and Machine learning from University College of London. A description of Michelangelo's research: As a PhD student at Queen Mary University of London, Michelangelo aims to leverage Bayesian models to develop principled algorithms for reinforcement learning in the context of function approximations. The main challenge lies in finding a balance between computational costs and optimality. Evaluating such balance requires careful evaluation, which is currently lacking in reinforcement learning. m.conserva@qmul.ac.uk Email Mastodon https://michelangeloconserva.github.io/ Other links Website https://www.linkedin.com/in/michelangeloconserva/ LinkedIn BlueSky https://github.com/MichelangeloConserva Github Supervisors: Prof. Simon Lucas Dr Paulo Rauber Featured Publication(s): What are you looking at? Team fight prediction through player camera Posterior Sampling for Deep Reinforcement Learning Hardness in Markov Decision Processes: Theory and Practice Recurrent Neural-Linear Posterior Sampling for Nonstationary Contextual Bandits The Graph Cut Kernel for Ranked Data Themes Game AI - Previous Next

< Back Dr Ryan Spick University of York iGGi Alum Deep Learning for Procedural Content Generation in Virtual Environments Ryan Spick is a PhD student with a computer science background, working on methods to improve how content (models, terrain, assets etc.) is created with an autonomous focus, with the main focus on generative deep learning to augment real-world data through a series of neural network layers to learn unlying properties of these data. Ryan has published a variety of papers around his main topic of generating content, such as terrain generation using generative adversarial networks and 3D voxel coloured model generation, to collaborations on other topics using deep learning, such as death prediction in a multiplayer online game and applying a recent map-elites algorithm. He has also worked with several leading industry researchers/games companies to further develop his research skill.If you have any ideas or collaboration opportunities please get in contact through any of the mediums below. Please note: Updating of profile text in progress ryan.spick@hotmail.co.uk Email Mastodon https://www.rjspick.com/ Other links Website https://www.linkedin.com/in/ryan-spick-505b63131/ LinkedIn BlueSky Github Featured Publication(s): System and Method for Point Cloud Generation System and method for training a machine learning model Robust Imitation Learning for Automated Game Testing Behavioural Cloning in VizDoom Illuminating Game Space Using MAP-Elites for Assisting Video Game Design Utilising VIPER for Parameter Space Exploration in Agent Based Wealth Distribution Models Human Point Cloud Generation using Deep Learning Naive mesh-to-mesh coloured model generation using 3D GANs Realistic and textured terrain generation using GANs Procedural Generation using Spatial GANs for Region-Specific Learning of Elevation Data Deep Learning for Wave Height Classification in Satellite Images for Offshore Wind Access Illuminating Game Space Using MAP-Elites for Assisting Video Game Design Time to die: Death prediction in dota 2 using deep learning Themes Game AI - Previous Next

< Back Nicole Levermore University of York iGGi PG Researcher Available for placement Nicole's academic background is within Neuroscience, having achieved BSc Neuroscience and Psychology, MSc Translational Neuroscience and an MPhil in Auditory Neuroscience. Outside of her research interests, she enjoys playing video games, hiking and playing the cello. A description of Nicole's research: Video games have enormous potential for research on cognition and mental health. In my project, I will use video games to perform basic research into a common psychiatric disorder (ADHD), paving the way for improved diagnosis, monitoring and therapy. ADHD is typically diagnosed in childhood and is characterised by failures of attentional state maintenance. This project involves using cutting-edge neuroimaging techniques to investigate how subjects with and without ADHD switch between attentional states (for example, ‘engagement’ and ‘flow’) while playing a cognitively engaging video game. The ultimate goal is to use video games to understand how mental health impacts people’s ability to focus on cognitively demanding tasks and, potentially, to develop therapeutic intervention. nicole.levermore@york.ac.uk Email Mastodon Other links Website https://www.linkedin.com/in/nicole-levermore-b14245283 LinkedIn BlueSky Github Supervisor: Prof. Alex Wade Themes Accessibility Design & Development Immersive Technology Player Research https://www.youtube.com/watch?v=gRFe1EOPW_4 Previous Next

< Back David Hull University of York iGGi Manager iGGi Admin I have worked at the University of York since October 1995, almost all of it in the Department of Computer Science. My various roles have included Laboratory and Facilities Manager, Technical Manager and, most recently, Project Manager. Outside work, I have been a change-ringer for almost 50 years, and am currently a member of the band that rings the bells weekly at York Minster. I am also an accredited teacher of bellringing. I do parkrun most weeks, alongside the occasional 10k and half marathon, like to watch cricket, and play the clarinet and piano. iggi-admin@york.ac.uk Email Mastodon Other links Website LinkedIn BlueSky Github Themes - Previous Next

< Back Dr Dan Franks University of York Supervisor Dr Franks is an interdisciplinary researcher and data scientist interested in AI and machine learning. He is experienced in developing and applying evolutionary computation and machine learning methods to understanding behaviour. He is an internationally recognized leader in interdisciplinary research, has published in top journals such as Science and PNAS. Some of his papers are in the top 1% of all papers for media coverage (altmetric), and his work is regularly covered by The New Scientist, National Geographic, Wired, The BBC, The Guardian, The Times, among others. As Reader in the York Centre for Cross-disciplinary Systems Analysis, Dan works on applying AI, machine learning, and agent-based modelling, to problems in other disciplines. Particular interests involve the development of machine learning methods for creating intelligent AI and for understanding complex systems. Research themes: Game AI Game Analytics daniel.franks@york.ac.uk Email Mastodon Other links Website LinkedIn BlueSky Github Themes Game AI Game Data - Previous Next

< Back Dr Cristina Guerrero-Romero Queen Mary University of London iGGi Alum Cris is a versatile Software Engineer with four years of experience in web development across different areas of the tech stack. She studied Software and Computer Engineering at Universidad Autónoma de Madrid (Spain) and is currently completing her PhD at Queen Mary University of London (QMUL); during which she has done two internships at Google. Her research ‘Beyond Playing to Win: Broadening the Study and Use of Gameplaying Agents when Provided with Distinct Behaviours’ is focused on expanding the research on game-playing agents beyond the objective of winning at them. She looks at 1) broadening the scope by diversifying agents goals and heuristics; 2) broadening the vision by proposing a team of agents to assist game development; 3) broadening the usage by eliciting diverse automated gameplay, and 4) broadening the horizon by analysing the strengths of the agents from a Player Experience perspective instead of their performance. Cris is passionate about solving problems and learning. Outside of her work, she enjoys playing video games and TTRPGs. Random facts are that Portal and TLOU are two of her favourite game series and her chosen superpower would be teleportation. Please note: Updating of profile text in progress Email Mastodon http://kisenshi.github.io/ Other links Website https://www.linkedin.com/in/cguerreromero/ LinkedIn BlueSky https://github.com/kisenshi Github Featured Publication(s): Beyond Playing to Win: Elicit General Gameplaying Agents with Distinct Behaviours to Assist Game Development and Testing Beyond Playing to Win: Creating a Team of Agents with Distinct Behaviours for Automated Gameplay MAP-Elites to Generate a Team of Agents that Elicits Diverse Automated Gameplay Generating Diverse and Competitive Play-Styles for Strategy Games Studying General Agents in Video Games from the Perspective of Player Experience Ensemble Decision Systems for General Video Game Playing Using a Team of General AI Algorithms to Assist Game Design and Testing Beyond playing to win: Diversifying heuristics for GVGAI Themes Design & Development Game AI - Previous Next

< Back Dr Cristina Dobre Goldsmiths iGGi Alum Cristina Dobre has a background in Mathematics and Computing receiving distinction in her undergraduate degree in Computer Science. My current focus is on the nonverbal cues that influence and shape the social interaction in immersive VR environments. More broadly, I'm investigating autonomous agents (or virtual humans) in social settings in terms of non-verbal interactions with users. I'm interested in the underlying mechanics of social interaction that help developing an emphatic and engaging virtual human. At the moment, I'm working on ML models based on multimodal datasets to detect various social cues (such as gaze) or various human-defined social attitudes (such as engagement) in social interactions in VR. I'm also interested in generating more complex behaviour for virtual characters (NPCs) that will improve the user's experience with the NPCs in a social VR setting. Designing communication and other social interactions in immersive VR can be a challenging task, and aspects on this are addressed in my research. The findings from these studies can help game designers and game developers determine the appropriate non-player character's non-verbal (and verbal) behaviour in games, especially in VR games. Along with its applications in the games industry, the findings would be useful for other applications such as designing multi-modal human-machine interactions and other systems for medical purposes, for social anxiety disorders therapy, simulations, training or learning. cristina.dobre@uni-a.de Email https://hci.social/@ShesCristina Mastodon Other links Website https://linkedin.com/shesCristina LinkedIn BlueSky https://www.github.com/shesCristina Github Featured Publication(s): Social Interactions in Immersive Virtual Environments: People, Agents, and Avatars Rolling Horizon Co-evolution in Two-player General Video Game Playing Using machine learning to generate engaging behaviours in immersive virtual environments More than buttons on controllers: engaging social interactions in narrative VR games through social attitudes detection Nice is Different than Good: Longitudinal Communicative Effects of Realistic and Cartoon Avatars in Real Mixed Reality Work Meetings Immersive Machine Learning for Social Attitude Detection in Virtual Reality Narrative Games Direct Gaze Triggers Higher Frequency of Gaze Change: An Automatic Analysis of Dyads in Unstructured Conversation Themes Game AI Immersive Technology - Previous Next

< Back Igor Dall'Avanzi Goldsmiths iGGi Alum Creation of accessible tools for the use of procedural audio in video games The aim of this research is to investigate and provide new tools to developers for the use of procedural audio into video games. Procedural approaches could address different issues that commonly afflict game audio. In music, generative systems are not only less repetitive, but offer more adaptability as well. For what concerns sound design, they can provide not only variety, but stronger and more realistic support to the interaction with the game world; interaction that is becoming even deeper with the advent of VR Yet, these methods still need improvement on different sides. One is the level of quality that procedural audio needs to achieve to compete with the current aesthetic established by the use of rendered sounds and music in the media. Another is the additional amount of work required by the CPU to render the assets on runtime, and its variable cost). Finally, there is a general lack of user-friendly tools, to link common programming languages for audio to game engines. Software like MaxMsp, Pure Data or SuperCollider is used to design generative audio systems. A more accessible integration of these software could promote generative approaches among sound designers and composers in the field, that today have instead access to tools mainly designed to be used with rendered assets. My plan is to bring on research first by focusing on how a higher degree of quality could be addressed, exploring tools like the above mentioned MaxMsp, Pure Data, low level solutions, and machine learning algorithms. Primary research will be run to confront procedurally generated audio content with rendered one; to understand its impact on the player, and the level of quality needed to deliver a satisfactory experience. The creation of more accessible interfaces and tools dedicated to implement procedural audio in video games will be investigated and undertaken. I like to make noises of all sort and to play with them. For this reason I graduated in Music Production in 2016 and, at the moment of writing, I am finishing my final project for an MSc in Sound and Music for Interactive Games at Leeds Beckett University. Composer and sound designer, in the last year I have been focusing on audio implementation and programming, and I am currently exploring machine learning approaches for procedural audio. Please note: Updating of profile text in progress Email Mastodon Other links Website LinkedIn BlueSky Github Themes Game Audio Player Research - Previous Next

< Back Dr Ben Kirman University of York iGGi Training Coordinator Supervisor Available to supervise non-iGGi students for 2024 intake Ben is a Senior Lecturer (Associate Prof) in Interactive Media at the University of York, who has over 20 years' experience as a creative technologist. Since his first programming job fixing Y2K bugs (you're welcome), he has worked with dozens of organisations, large and small, in design and prototyping playful experiences. His research uses game design and playful design as a way to explore the complex effects of emerging technologies through novel and unexpected interactions and experiences. Most often, this is through the design and development of games, digital/physical prototypes, and design fictions. Ben has applied this in topics ranging from immersive theatre, dog technology, non-league football, radical cycle delivery, and time travelling robots, to educational games, esports, new situationism and magic. The unifying theme is play – as a topic of study, a way of working, for research insight, and as expression or output in games or playful experiences. This work, especially the more bizarre stuff, has often been covered by traditional media, including the BBC, New Scientist, Wired, The Guardian, TIME, Metro, the New York Times, and Your Cat magazine. Ben is keen on supervising students with strong creative drives, with an interest in making, design, experimentation, and a broad perspective on games and play. This might be a project about playful props in immersive theatre, or a project about context in locative and site-specific games, or any other project that looks to explore new possibilities and new implications of emerging technology through the lens of play. Research themes include: Game Design Applied Games Computational Creativity Sports with an E and without an E Player Experience ben.kirman@york.ac.uk Email Mastodon https://ben.kirman.org/ Other links Website LinkedIn BlueSky Github Themes Applied Games Creative Computing Design & Development Esports Player Research - Previous Next

< Back Dr Paulo Rauber Queen Mary University of London Supervisor I am a lecturer in Artificial Intelligence at Queen Mary University of London. Before becoming a lecturer, I was a postdoctoral researcher in the Swiss AI lab working on reinforcement learning under the supervision of Jürgen Schmidhuber. I believe that intelligence should be defined as a measure of the ability of an agent to achieve goals in a wide range of environments, which makes reinforcement learning an excellent framework to study many challenges that intelligent agents are bound to face. p.rauber@qmul.ac.uk Email Mastodon https://paulorauber.com/ Other links Website LinkedIn BlueSky https://github.com/paulorauber Github Themes Game AI - Previous Next

< Back Filip Sroka Queen Mary University of London iGGi PG Researcher Filip is a Computer Science researcher specialising in Game AI. He acquired an Integrated Masters in Computer Science from Queen Mary University of London and is pursuing a PhD in Game AI with iGGi. With a passion for algorithms and problem-solving, he constantly seeks new challenges to enhance his skills. As an avid LEGO collector and investor, he brings a unique blend of technical and creative abilities. He is excited about the potential of the Metaverse and is driven by the role of technology in shaping its future. His research explores the integration of Dynamic Difficulty Adjustment (DDA) into VR rhythm games such as Beat Saber, with the goal of enhancing player skill development and motivation through the application of learning theories. By addressing difficulty spikes, the project creates personalised learning experiences using human-made maps designed to accelerate the learning process. Key components include player evaluation, map segmentation, and procedural generation. The broader aim is to extend these findings to other rhythm games, offering benefits to players, game developers, and the health and fitness industry. f.sroka@qmul.ac.uk Email Mastodon Other links Website https://www.linkedin.com/in/filip-sroka-134954197/ LinkedIn BlueSky https://github.com/FilipSroka Github Supervisor: Dr Laurissa Tokarchuk Themes Applied Games Game AI Immersive Technology - Previous Next

< Back Prof. Richard Bartle University of Essex iGGi Co-Investigator Supervisor Richard Bartle is a renowned pioneer in game design and research. He co-wrote the first virtual world, MUD ("Multi-User Dungeon") in 1978, and has thus been at the forefront of the online games industry from its very inception. He is an influential writer on all aspects of virtual world design, development, and management. As an independent consultant, he has worked with many of the major online game companies in the U.K. and the U.S. over the past 30 years. His 2003 book, Designing Virtual Worlds , has established itself as a foundation text for researchers and developers of virtual worlds alike. His Player Type theory is taught in game design programmes worldwide (he appears in examination questions!). His interests are directed mainly virtual worlds, particularly Massively Multiplayer Online Role-Playing Games (MMORPGs, or MMOs), but cover all aspects of game design. He is keen to see AI used for non-player characters in MMOs (his PhD is in AI), and his current work considers the long-term moral and ethical implications of this. They’re maybe not what you might think they were at first glance… rabartle@essex.ac.uk Email Mastodon https://mud.co.uk/richard/ Other links Website https://www.linkedin.com/in/richardbartle/ LinkedIn BlueSky Github Themes Design & Development Game AI Player Research - Previous Next