Search Results

< Back Dr Sebastian Berns Queen Mary University of London iGGi Alum Sebastian is a designer and researcher working on use-inspired fundamental research in generative machine learning for creative and artistic applications. Sebastian holds a master’s degree in artificial intelligence and has a background in visual communications. He has worked several years as an independent graphic and type designer with a specialisation in web development. His design work has been awarded national and international design prizes. A description of Sebastian's research: "Generative machine learning methods are trained on raw data, modelling the primary patterns that constitute typical examples. They enable the production of high-quality artefacts in very complex domains and provide useful models for generative systems, in particular in the visual arts and video games. However, modelling a training data distribution perfectly is less valuable for applications in art production and video games. In particular, our analysis of the use of generative models in visual art practices motivates the need to increase the output diversity of generative models. In my dissertation, I focus on diversity in generative machine learning for visual arts and video games. Our findings benefit the application of generative models in generative systems, quality diversity search, art production and video games. Rather than a ‘ground truth’ that needs to be modelled perfectly, we argue that training datasets are merely a limited snapshot of a complex world with inherent biases. To be useful for applications in visual arts and video games, generative models require higher output diversity. Relatedly, higher generative diversity benefits efforts of equity, diversity and inclusion by reducing harmful biases in generative models." s.berns@qmul.ac.uk Email Mastodon http://www.sebastianberns.com/ Other links Website LinkedIn BlueSky https://github.com/sebastianberns Github Featured Publication(s): Not All the Same: Understanding and Informing Similarity Estimation in Tile-Based Video Games Towards Mode Balancing of Generative Models via Diversity Weights Increasing the Diversity of Deep Generative Models Active Divergence with Generative Deep Learning--A Survey and Taxonomy Automating Generative Deep Learning for Artistic Purposes: Challenges and Opportunities Expressivity of Parameterized and Data-driven Representations in Quality Diversity Search First experiments in the automatic generation of pseudo-profound pseudo-bullshit image titles Generative Search Engines: Initial Experiments Adapting and Enhancing Evolutionary Art for Casual Creation. Creativity Theatre for Demonstrable Computational Creativity Bridging Generative Deep Learning and Computational Creativity NEST 2.18. 0 Active Divergence with Generative Deep Learning--A Survey and Taxonomy Themes Creative Computing - Previous Next

< Back Dr Shringi Kumari University of York iGGi Alum Shringi is a seasoned game designer with more than nine years of experience making games for companies including EA, Zynga, Bigpoint, and Wooga. She became a researcher four years ago, wondering how game designers can take inspiration from other creative fields. In her PhD, she is now studying how stage magic can be translated to games for creating believable illusions of choice and moments of surprise. She continues to consult as a game designer for companies and has started a lecturership in game design at University of East London. In the past years she has spoken about game design across the world at a number of known platforms: Indiecade Europe, Develop, Game Happens, SOMA Chicago, GDC India to count some. As a creative, she engages in working on disruptive design both in games and beyond. Her work reflects her Indian background and discusses universal issues of identity, need for diversity and the idea or illusion of home. She has recently published her debut poetry collection,“The Saree Shop” and has featured in a short story anthology with her story ”Garden of Vaginas”. Shringi is supervised by Dr Sebastian Deterding (York) and Dr Gustav Kuhn (Goldsmiths). Please note: Updating of profile text in progress Email Mastodon https://shringikumari.com Other links Website https://www.linkedin.com/in/shringi-kumari-8613678 LinkedIn BlueSky Github Featured Publication(s): The role of uncertainty in moment-to-moment player motivation: a grounded theory Why game designers should study magic Investigating uncertainty in digital games and its impact on player immersion Studying General Agents in Video Games from the Perspective of Player Experience The Magician's Choice: Providing illusory choice and sense of agency with the Equivoque forcing technique. Design Inspiration for Motivating Uncertainty in Games using Stage Magic Principles Themes Player Research - Previous Next

< Back Oliver Roughton University of York iGGi Administrator iGGi Admin Based in York alongside Tracy and Helen I act as a Point of contact for iGGi PGRs and provide administrative support in the implementation of iGGi procedures. iGGi PGRs are most likely to hear from me in relation to conference/kit funding and travel bookings for the taught modules and other iGGi events. As well as my admin work I am a part-time PhD student (not with iGGi) and spend much of my free time knitting. oliver.roughton@york.ac.uk Email https://www.instagram.com/klaus.the.magnificent/ Mastodon Other links Website LinkedIn BlueSky Github Themes Previous Next

< Back Prof. Greg Slabaugh Queen Mary University of London Supervisor Gregory G. Slabaugh is Professor of Computer Vision and AI and Director of the Digital Environment Research Institute (DERI) at Queen Mary University of London. He is also a Turing Fellow at the Alan Turing Institute. His research work spans computer vision and computer graphics including geometric modelling and image/video-based understanding. He is interested in deep learning approaches including generative techniques like normalizing flow an generative adversarial networks. He previously worked in the games industry as a 3D graphics programmer and his PhD thesis focussed on how to model 3D objects from a collection of images. He is interested in how to create engaging content and interaction from images as well as procedural methods to reduce the effort of 3D modelling. g.slabaugh@qmul.ac.uk Email Mastodon https://www.eecs.qmul.ac.uk/~gslabaugh Other links Website https://www.linkedin.com/in/greg-slabaugh-a5b03a1/ LinkedIn BlueSky Github Themes Applied Games Creative Computing Immersive Technology - Previous Next

< Back Prof. Nick Pears University of York Supervisor Nick Pears is a Professor of Computer Vision in York’s Vision, Graphics and Learning (VGL) research group. He works on statistical modelling of 3D shapes, with an emphasis on the human face and head. The Liverpool-York Head Model and the associated Headspace training set has been downloaded by over 100 research groups internationally, with the Universal Head Model being downloaded by 50 research groups. His most recent work with his PhD students has focused on semantic disentanglement of 3D images and how to make autonomous vehicles safer and more trustworthy when using computer vision systems. He is assessor for many PhDs including construction of generative models for novel video content using adversarial deep learning techniques. nick.pears@york.ac.uk Email Mastodon https://www-users.cs.york.ac.uk/np7/ Other links Website https://www.linkedin.com/in/nick-pears-90970312/ LinkedIn BlueSky Github Themes Creative Computing Game AI - Previous Next

< Back Dr Daniel Hernandez University of York iGGi Alum With the games industry as his target, Daniel Hernandez’s main research objective is to design and implement algorithms that, without any prior knowledge, generate strong gameplaying agents for a wide variety of games. To tackle this “from scratch” learning, he uses, and contributes to, the fields of Multiagent Reinforcement Learning, Game Theory and Deep learning. Self-play is the main object of study in his research. Self-play is a training scheme for multiagent systems in which AIs are trained by acting on an environment against themselves or previous versions of themselves. Such training scheme bypasses obstacles faced by many other training approaches which rely on existing datasets of expert moves or human / AI agents to train against. Daniel’s hope is that further development in Self-play will allow game studios of all sizes to generate strong AI agents for their games in an affordable manner. A storyteller by nature, Daniel has a strong track record of outreach through talks and workshops both in the UK and internationally. By sharing his journey, insights and discoveries he hopes to both inspire and instruct students, researchers and developers to realise the potential that Reinforcement Learning has to improve the games industry. His passionate work on Machine learning goes beyond crafting strong gameplaying agents. He sees the potential of using AI to simplify and automate a wide range of tasks in the games industry. He has led successful projects which used machine learning aimed at automating multiagent game balancing to alleviate the burden of manual game balancing. Daniel received an MEng in Computing: Games, Vision & Interaction from Imperial College London. Wanting to combine the power of AI and the creativity of videogames, Daniel began a PhD journey to explore the misty lands of Multi Agent Reinforcement Learning (MARL). Please note: Updating of profile text in progress Email Mastodon https://danielhp95.github.io Other links Website https://www.linkedin.com/in/dani-hernandez-perez-1106b2107 LinkedIn BlueSky https://github.com/Danielhp95 Github Featured Publication(s): A comparison of self-play algorithms under a generalized framework A generalized framework for self-play training Metagame Autobalancing for Competitive Multiplayer Games Themes Game AI Player Research - Previous Next

< Back Dr Nuria Peña Pérez Queen Mary University of London iGGi Alum Nuria got her bachelor’s in biomedical engineering in Spain before moving to London. After studying an MSc in Neurotechnology and working in robotic neurorehabilitation at Imperial College London, she discovered the enormous potential of serious games in the field of human-robot interaction. She joined IGGI in 2018. Her PhD research involves studying human motor control and learning during bimanual tasks to investigate how the dynamics of the interaction can serve to develop better training systems. This is done through the development of interactive gaming environments that are compatible with rehabilitation robotic devices. The modelling of the recorded human neuromuscular data allows to explore how to better help patients to restore their motor function. Her work is a collaboration between the Advanced Robotics group at Queen Mary University of London and the Human Robotics group at Imperial College London. As part of her PhD she has worked for the company GripAble, developing games for the assessment and training of hand function (February 2020-August-2020). n.penaperez@qmul.ac.uk Email Mastodon Other links Website LinkedIn BlueSky Github Supervisor(s): Dr Ildar Farkhatdinov Featured Publication(s): Redundancy Resolution in Trimanual vs. Bimanual Tracking Tasks Dissociating haptic feedback from physical assistance does not improve motor performance Bimanual interaction in virtually and mechanically coupled tasks The impact of stiffness in bimanual versus dyadic interactions requiring force exchange How virtual and mechanical coupling impact bimanual tracking Lateralization of impedance control in dynamic versus static bimanual tasks Is a robot needed to modify human effort in bimanual tracking? Exploring user motor behaviour in bimanual interactive video games Quartz Crystal Resonator for Real-Time Characterization of Nanoscale Phenomena Relevant for Biomedical Applications Illuminating Game Space Using MAP-Elites for Assisting Video Game Design Themes Applied Games - Previous Next

< Back Dr Tom Cole iGGi Alum + Supervisor Games should be studied as interactive systems, but are more often studied using techniques reserved for non-interactive media. As developers, we are ‘selling ourselves short’, and not exploring the creative and expressive potential of digital games to their fullest. Out of the myriad of affective experiences possible, we generally only design and experience a fraction of what could be offered. Tom hopes to help address this by studying how game mechanics, gameplay systems and control methods can be used and interpreted to create meaning and elicit a wider range of emotional responses than is commonly seen in digital games at present. Broadening and deepening emotional engagement with an emphasis on mechanics and systems. (Industry placement at Bossa Studios) Video games, with their unique properties such as interactivity, agency, control mechanics, feedback loops and gameplay systems, have the potential to impart deep emotional experiences – some already do of course. However, study of this emotional engagement remains lacking. Reliance on techniques and theory appropriated from film, literature and cultural studies yields limited results. There is relatively little understanding of how procedural elements such as control mechanisms and gameplay systems can be leveraged (or synergised with narrative and/or audio-visual elements) for emotional affect. Tom was previously at Supermassive Games where he was a designer on the BAFTA award-winning horror game Until Dawn and artist on Killzone Shadow Fall. Tom got his BSc in Biology with Industrial Experience from Manchester. After teaching science in secondary schools for a while, he decided games were more interesting and got his MA in Digital Games Theory and Design at Brunel. After time at Goldsmiths, University of London and the University for Creative Arts, Rochester, Tom is now Lecturer in Games Development at the University of Greenwich where he teaches games development, design and production. From 2016 to 2024 he led the organisation of Adventurex - the Narrative Games Convention, a sold out international conference which grew from 100 to 650 people during his time leading it. tom@tommakesgames.com Email Mastodon http://www.tommakesgames.com Other links Website https://uk.linkedin.com/in/tom-cole-87043a38 LinkedIn BlueSky Github Featured Publication(s): Emotional exploration and the eudaimonic gameplay experience: A grounded theory More than a bit of coding:(un-) Grounded (non-) Theory in HCI Eudaimonia in Digital Games Thinking and doing: Challenge, agency, and the eudaimonic experience in video games "Moments to Talk About": Designing for the Eudaimonic Gameplay Experience Grounded Theory in games research: making the case and exploring the options Grounded Theory in Games Research: Making the Case and Exploring the Options Emotional and functional challenge in core and avant-garde games The Tragedy of Betrayal: How the design of Ico and Shadow of the Colossus elicits emotion Themes Design & Development Game AI https://www.youtube.com/playlist?feature=share&list=PL_17c-ELEJ5334QRqxhRLnnoX8aNdpHL- - https://www.youtube.com/watch?v=8pe5FfHTk-4 Previous Next

< Back Dr Søren Riis Queen Mary University of London Supervisor Søren Riis has more than 15 years of experience in teaching computability, complexity and the art of creating fast efficient algorithms. He has a strong interest in reinforcement learning and generative adversarial networks (GANs) related to strategy games. Riis has been actively involved in computer chess, and is listed on the wiki of influential people in chess programming https://www.chessprogramming.org/ Søren Riis is a strong player of strategy games including Chess, Shogi, Go and Bridge at an internal level. He has worked as a consultant for an AI company and is involved in applying deep learning for the card game of bridge. For the last 5 years he has been working on technical projects related to machine learning and reinforcement learning. He has practical experience and interest in scientific computing on super computers, and in creating C and C++ libraries to run from within python. Søren Riis is particularly interested in supervising students with a strong technical and/or maths background. Aptitude for strategy games with an interest in one the following ares is an advantage. Games requiring inductive reasoning combined with exploration. Hidden identity games (Werewolf, Resistance/Avalon, Mafia etc) Using GANs to sample realistic scenarios during gameplay Deep Reinforcement Learning in multi-agent strategy games Building and analysing games for investigating evolution of communication. Research themes: Game AI Game Design Game Creativity Games and mathematics s.riis@qmul.ac.uk Email Mastodon https://www.eecs.qmul.ac.uk/profiles/riissoren.html Other links Website https://www.linkedin.com/soren-riis-13602117/ LinkedIn BlueSky Github Themes Creative Computing Game AI Game Data - Previous Next

< Back Peyman Hosseini Queen Mary University of London iGGi PG Researcher Peyman is interested in using his computer science knowledge to support society's well-being. Raised in a family where almost everyone’s work is somehow related to mathematics and its applications, he became passionate about algorithms and combinatorics from an early age. This prompted him to pursue an undergraduate degree in computer engineering with a focus on IT and AI. This background led him to start his PhD at IGGI on building more powerful yet efficient Natural Language Processing models for analysing textual data, a rich and abundant source of gaming feedback. A description of Peyman's research: Peyman's research focuses on advancing deep learning architectures for natural language processing and building tools on top of state-of-the-art models. To contribute to the fundamental understanding and practical application of deep learning in natural language processing, focusing on efficiency and effectiveness, he pursues two main objectives: Designing more efficient models that match or surpass state-of-the-art performance with fewer parameters. Systematically analyzing language models to develop solutions that enhance their effectiveness for end-users, such as game studios. His recent accomplishments towards these goals include: 1. Developing novel attention mechanisms: 1.1 Optimized Attention: 25% parameter reduction 1.2 Efficient Attention: 50% parameter reduction 1.3 Super Attention: 25% parameter reduction with significant performance improvements in language and vision tasks 1.4 All mechanisms demonstrate comparable or superior performance to standard attention across various inputs. 2. Designing and training Hummingbird , a proof-of-concept small language model using Efficient Attention, available on HuggingFace. 3. Conducting a study on large language models' limitations in analyzing lengthy reviews for basic NLP tasks. Proposed solutions offer substantial performance improvements while reducing API costs by more than 90%. s.hosseini@qmul.ac.uk Email Mastodon https://peymanhosseini.net/ Other links Website https://www.linkedin.com/in/peyman-hosseini1 LinkedIn BlueSky https://github.com/Speymanhs Github Supervisors: Dr Ignacio Castro Prof. Matthew Purver Featured Publication(s): Cost-Effective Attention Mechanisms for Low Resource Settings: Necessity & Sufficiency of Linear Transformations Efficient solutions for an intriguing failure of llms: Long context window does not mean LLMs can analyze long sequences flawlessly Brain Drain Optimization (BRADO) Algorithm to Solve Multi-Objective Expert Team Formation Problem in Social Networks You Need to Pay Better Attention: Rethinking the Mathematics of Attention Mechanism GeoPos: A Minimal Positional Encoding for Enhanced Fine-Grained Details in Image Synthesis Using Convolutional Neural Networks Lon-ea at SemEval-2023 Task 11: A Comparison of Activation Functions for Soft and Hard Label Prediction GRACER: Improving the Accuracy of RACER Classifier Using A Greedy Approach Themes Game AI Player Research - Previous Next

< Back Rokas Volkovas Queen Mary University of London iGGi Alum Application of Neuroevolution to General Video Game Playing In the field of artificial intelligence, great advancements in developing AI capable of playing specific games has been made over last few decades. Over the years, the potential of General Game Playing (GGP) AI, was realized, and thus a new area of research was spawned, focusing mainly on turn-based board games. Rapidly expanding, it was just recently extended to include video games and has morphed into General Video Game Playing (GVGP). The studies in this space of AI are highly attractive due to their solution capacity of being highly transferable. As the field is relatively new, there are many different paths to explore. Some effort has already been put into incorporating the established Genetic Algorithm techniques into the area. The goal of the proposed research is to further develop models using the more complex evolutionary algorithms to find generalist solutions to the problems exposed in GVGP. More specifically, the research will aim to discover the appropriate applications and the modifications necessary of approaches such as Competitive Coevolution, circumventing its drawbacks and evolving populations capable of playing multiple games. Furthermore, in addition to other methods it will be concerned with the application of models developing generalist memory on a slower scale evolution (compared to individual in a population) with continuous state perturbations, to find closer to optimum results - adapting networks of individuals to the fitness landscape. In order to reach the goals of the research a number of experiments will be conducted, using a select few video games as a base performance measure. Training the populations evolved will involve tuning the evolutionary operators as well as altering pre-designed system be- haviours to suitably compare the viability of applied procedures. The success of bridging EA with GVPG, along with its advantages and drawbacks in the field will be readily deter- mined, comparing the solutions found to those of other existing approaches. Specifically, the similarity of the behaviour in evolvability using genetic networks searching for solutions and learning theory, via neural networks, has recently been suggested. Evolution is defined to not have any foresight, but models were built showing how it can remember previously discovered solutions, which would imply that natural selection leans towards long term evolvability. Kostas Kouvaris et. al. further establishes the underlying equivalence of the approaches, applying machine learning techniques to improve the generalisation of EA. The generalization allows combining the features from previous experience to find individuals with new feature combinations, better adapted to unseen environments. Were the exploratory learning methods developed in EA to perform no less satisfactorily in the gaming industry environment, given enough sample data from a handful of well defined behaviours, the AI units could be trained to adapt to the new levels they are placed in. In theory, this would then translate to the same amount of effort producing a larger variety of content or, alternatively, producing the same amount of content with less effort, distributing the excess to other areas of development or eliminating it to lower the total production cost. Rokas is an MEng Electronic Engineering graduate from University of Southampton. Initially, pushed away from programming in school due to being taught Pascal, he realized its power in the compulsory C course in University. Applying the knowledge to building games caused the gradual shift from electronics to software development, with the 4th year modules all having the CS tag. During the undergraduate studies Rokas held the UKESF scholarship and did 2 summer internships at Imagination Technologies. Interests in game and software development got him researching neuroevolutionary machine learning for video games. Please note: Updating of profile text in progress Email Mastodon Other links Website LinkedIn BlueSky Github Featured Publication(s): Automatic Game Tuning for Strategic Diversity Practical Game Design Tool: State Explorer Extracting learning curves from puzzle games Mek: Mechanics prototyping tool for 2d tile-based turn-based deterministic games Diversity maintenance using a population of repelling random-mutation hill climbers Themes Game AI - Previous Next

< Back Dr Tom Collins University of York Supervisor Tom runs the Music Computing and Psychology Lab in the Music Department at University of York, and so makes a good supervisor for game audio projects, but he has wider interests in media (e.g., podcasts) and sport (especially football), and in sport how AI can be leveraged to enhance analytics that lead to new insights into, and competitive advantages in, individual and team performance. Tom is internationally recognised for his work in automatic music generation, web systems for music, and information retrieval. His research has been featured by the BBC (BBC Click), The Times, and Financial Times among others. Tom is interested in supervising students who have a background in at least one of the following areas, and who are interested in acquiring knowledge of the others: Data science and machine learning (especially deep learning); One of music, podcasts, or sport; Software engineering (especially full-stack JavaScript development). tom.collins@york.ac.uk Email Mastodon https://tomcollinsresearch.net Other links Website LinkedIn BlueSky Github Themes Esports Game AI Game Audio Game Data Player Research - Previous Next