Search Results
Search this site
Results found for empty search
- Peyman Hosseini
< Back Peyman Hosseini Queen Mary University of London iGGi PG Researcher Peyman Hosseini is a PhD candidate working on Agentic AI and building on efficient solutions with small language models and post-training of large and small language models to enable these LLMs to be powerful on-device assistants. He has interned for the last 12 months at Samsung Research in the UK where he has led 3 paper sumbissions and 2 patent submissions on post training foundation models with reinforcement learning algorithms as well as building efficient on-device memory agents. A description of Peyman's research: Peyman's Rsearch targets post-training of foundation models, specifically large language models, to deliver personalized and powerful AI-powered solutions that are deployable on edge devices, such as mobile phones and personal computers. This is specifically important as Large Language Models (LLMs) are powerful yet impossible to deploy on edge-devices to their computational requirements. On the other hand, Small Language Models (SLMs), i.e., language models between 2-32B params, are more efficient but yet unable to handle complex tasks. Fine-tuning these models to work well in complicated setting enables a lot powerful, privacy-preserving AI-powered applicatios, such as personalized on-device recommendation systems and agents capable of memorizing users' habits and interests. Email s.hosseini@qmul.ac.uk Website LinkedIn Mastodon BlueSky GitHub Other Link Supervisors: Dr Ignacio Castro Prof. Matthew Purver Featured Publication(s): DuoMem: Towards Capable On-Device Memory Agents via Dual-Space Distillation CG-TTRL: Context-Guided Test-Time Reinforcement Learning for On-Device Large Language Models 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
- Philip Smith
< Back Philip Smith Queen Mary University of London iGGi PG Researcher Available for placement I was born and raised in Bermuda, a small island in the Atlantic Ocean with an approximate population of 65,000 people. I finished my undergraduate degree in Computer Science with a Specialist in Game Design at the University of Toronto. For my Master's degree, I studied Computer Games Technology at City, University of London. My goal is to help expand the use of video games from purely recreational activities to viable avenues for aiding in real world problems. A description of Philip's research: My research will be focusing on maximizing player engagement in gamified citizen science as a continuation of my Master's thesis. 'Citizen science' is the practice of employing volunteers from the general public for the collection and/or processing of data with respect to a scientific project. Gamified citizen science projects have relied upon prolonged engagement from volunteers, but the number of long-term participants have been unsatisfactory in current projects. This project attempts to address the lack of sufficient volunteer engagement in gamified citizen science projects. The aim is to build a framework meant to guide game designers in creating an engaging citizen science video game based on the values set by Self-Determination Theory (SDT). These values adhere to the theory’s concept of intrinsic and extrinsic motivators of engagement. Intrinsic motivation relies on the factors of player autonomy, competence, and relatedness during gameplay. Extrinsic motivation relies on external incentives to core gameplay such as in-game rewards. As part of my research, I am evaluating multiple game design frameworks focused on Applied Games and identifying the merits and flaws each have when applied to a citizen science context. The information I gather will formulate a prototype of the Framework that will be iterated upon through design workshops, development, and playtesting. Email p.c.smithii@qmul.ac.uk Website LinkedIn Mastodon BlueSky GitHub Other Link Themes Applied Games Design & Development - Previous Next
- Filip Sroka
< Back Filip Sroka Queen Mary University of London iGGi PG Researcher Available for placement 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. An avid LEGO collector and investor, Filip brings a unique blend of technical and creative abilities to his work. He is excited about the potential of the Metaverse and is driven by the role of technology in shaping its future. A description of Filip's research: His research explores the integration of Dynamic Difficulty Adjustment (DDA) and Procedural Content Generation (PCG) into VR rhythm games to optimise motor learning and skill acquisition. By leveraging learning theories, the project creates personalised, adaptive training environments. Beyond commercial gaming, this framework demonstrates how adaptive systems can be used to maximise engagement and training efficiency, offering high-value insights for game developers, immersive tech creators, and the digital health and fitness industry. Email f.sroka@qmul.ac.uk Website LinkedIn Mastodon BlueSky GitHub Other Link Supervisors: Dr Alena Denisova Dr Laurissa Tokarchuk Dr Jeremy Gow Themes Applied Games Game AI Immersive Technology - Previous Next
- Dr Cade McCall
< Back Dr Cade McCall University of York Supervisor Cade McCall is an experimental psychologist. He uses games and virtual environments to study emotion, cognition, and behaviour during threatening experiences. His work explores how threat unfolds over time as revealed by dynamics in motion tracking data, psychophysiological measures, and experience-sampling. McCall is interested in supervising projects with a psychological focus, including: ● human interactions with autonomous systems ● the use of games to manipulate emotions ● social interactions within games Research themes: Games with a purpose Player experience Game analytics Email cade.mccall@york.ac.uk Website LinkedIn Mastodon BlueSky GitHub Other Link Themes Applied Games Game Data Player Research - Previous Next
- Dr David Zendle
< Back Dr David Zendle University of York Supervisor David Zendle is an active researcher into the effects of both video games and gambling, and is the author of several key references on the topic of video game monetisation. His most well-known publications deal with the potential effects of loot boxes. His recent work focuses on understanding the diversity of ways that video game play impacts wellbeing, and involves the analysis of large-scale datasets of player behaviour and spending. David is an academic affiliate of the Behavioural Insights Team and holds a research position within the NHS. He is particularly interested in building evidence-based policy in the domain of video game regulation, and has provided oral testimony on video game effects to a variety of government investigations across the globe. David is particularly interested in supervising students with an industry, economics, legal, or behavioural sciences background. He is interested in work on the following topics: The long-term effects of video game play (both positive and negative) Video game monetisation Video game regulation and policy Dark video game design Research themes: Game Analytics Game effects Game policy Email david.zendle@york.ac.uk Website LinkedIn Mastodon BlueSky GitHub Other Link Themes Game Data Player Research - Previous Next
- Joseph Walton-Rivers
< Back Dr Joseph Walton-Rivers University of Essex iGGi Alum Controlling Non-player characters. (Industry placement at Visteon) Within games non-player characters help to sell the world and give meaning to the player's experiences. These characters in games are presently not very believable and often lack the ability to interact with each other in meaningful ways. This work is looking at creating socially capable, believable agents to populate the worlds of role playing games. These agents need to be able to cope with player's actions and be capable of acting independent in the world. Joseph studied computer science at the University of Essex, obtaining a first class degree. During his study there he received two awards for academic achievement. After graduation he worked in the IT team of a company with offices across the United Kingdom where he developed and maintained their IT systems. Since starting IGGI he has worked on research involving co-operative agents working together to solve shared goals. He has a keen interest in programming and the Free Software movement. During his free time he enjoys strategy and puzzle games including Prison Architect, the Shadowrun series and Galactic Civilization 2. Please note: Updating of profile text in progress Email Website LinkedIn Mastodon BlueSky GitHub Other Link Featured Publication(s): Enhancing Programming Learning with AI-Generated Contextual Examples in Digital Creativity Peer Evaluation in Group Projects: Insight into Effective Student Critique and Feedback Evaluating the Effects on Monte Carlo Tree Search of Predicting Co-operative Agent Behaviour An Exploratory Analysis of Student Experiences with Peer Evaluation in Group Game Development Projects Student Perspectives on the Purpose of Peer Evaluation During Group Game Development Projects The 2018 Hanabi competition Hexboard: A generic game framework for turn-based strategy games Fireworks agent competition Evaluating and Modelling Hanabi-Playing Agents Controlling co-incidental non-player characters Monte carlo tree search applied to co-operative problems Distributed Social Multi-Agent Negotiation Framework For Incomplete Information Games Themes Player Research - Previous Next
- Francesca Foffano
< Back Francesca Foffano University of York iGGi PG Researcher Available for post-PhD position Francesca is finalising her PhD at the University of York, specialising in UX, player experience and design. Their work spans game experience (e.g. getting stuck), safeguarding the design's impact on players (e.g. emotional accessibility), design recommendations (e.g. emotional design) and game development (e.g. adaptive games). They primarily use qualitative research and mixed methods to gain an in-depth understanding of the player experience. They are currently open to collaborations with industry and academia for funding applications. A description of Francesca's research: When it comes to emotions in video games, players should feel like they are enthusiastically part of the experience. However, at times, players risk feeling forced or overwhelmed. So, how to design these experiences correctly? Through a dual approach, I explore how to design emotional experiences at their extremes. On the one hand, I examine which design components create recognisable emotional experiences. On the other hand, I examine how design can overwhelm players to the point where they struggle to finish a game, as the design makes it impossible for them to process the emotions presented. Email ff716@york.ac.uk Website LinkedIn Mastodon BlueSky GitHub Other Link Supervisor: Prof. Paul Cairns Featured Publication(s): Artificial intelligence across europe: A study on awareness, attitude and trust Disability tax in gacha games: the social cost of hidden expenses. Do Not Play With My Emotions. Design for Emotional Accessibility A Survey on AI and Ethics: Key factors in building AI trust and awareness across European citizens. When Games Become Inaccessible: A Constructive Grounded Theory on Stuckness in Videogames Artificial intelligence across europe: A study on awareness, attitude and trust Investing in AI for social good: an analysis of European national strategies European Strategy on AI: Are we truly fostering social good? Changes of user experience in an adaptive game: a study of an AI manager Themes Player Research - https://www.youtube.com/watch?v=v8xdnkjVL2c Previous Next
- Emily Marriott
< Back Emily Marriott University of Essex iGGi Alum Automated Story Generation for Games Emily is researching automated story generation for video games, focusing on the use of Planning for real-time, dynamic generation. Ideally, the stories created will reflect choices made by the player during gameplay and will update continually throughout gameplay. The aim of this research is to create a system that could be easily utilised in the development of more adaptive games. This could improve player enjoyment, increase re-playability, and allow for the inclusion or exclusion of content that may only appeal to niche audiences. Emily’s current focus is on investigating story structures and pacing to create a template for generating good stories specifically for games that are consistent, well-structured and interesting. This involves studying the pacing requirements in existing games to establish what these are and how they differ the requirements for film and TV. The system will ideally be integrated with existing game-development tools and provide an easy-to-use interface to make the creation of adaptive games easier and quicker. The eventual goal is a full story-generation system would support both the creation of quests that emerge from story requirements and a game world that fits the environment required for the story. Emily graduated from Glyndŵr University with a BSc in Computer Games Development before completing an MSc in Computer Science at Oxford Brookes University. The substance of the MSc dissertation involved generating dungeon levels and quests using grammars based on the play style the player appeared to favour. Emily enjoys playing both tabletop and computer roleplaying games, especially ones in which player actions can have a dramatic effect on the game’s progression. Please note: Updating of profile text in progress Email Website LinkedIn Mastodon BlueSky GitHub Other Link Themes Player Research - Previous Next
- Michael Aichmueller
< Back Michael Aichmüller Queen Mary University of London iGGi Alum My background lies in physics and statistical mathematics with a later specialization in optimization in the fields of Reinforcement Learning (RL) and Causal Inference. My first encounters with RL occurred during my Masters when studying how to create strong policies in perfect information games using algorithms, such as MinMax, MCTS, DQN, and later AlphaZero variants. My favorite game application remains the board game ‘Stratego’. In the meantime I investigated the estimation of causal parents influencing a target variable from interventional datasets for my Master’s thesis. Specifically, how well Deep Learning estimations could replace exponentially scaling graph search methods with approximations requiring only polynomial runtime. A description of Michael's research: My research focuses on the state-of-the-art in game-playing solutions for imperfect information games (think games like Poker, Stratego, Liar’s Dice etc.). I am particularly interested in the application of No-Regret (and related) methods which seek to learn those actions that provided the most benefit (or least regret) compared to the benefit all possible actions provided on average. These methods learn such via iterative play to find a Nash-Equilibrium (NE), a game-theoretic concept comparable to an optimal policy known from Single-Agent RL, but for all partaking players at once. Particularly, variants of Counterfactual Regret Minimization (CFR) remain the state-of-the-art algorithms for computing NEs in 2-player zero-sum games due to their success in tabular form so far. Yet, prohibitive complexity and memory scaling bars them from large-scale applications. Hence, research of recent years seeks to couple CFR (and other No-Regret methods) with function approximation, such as Deep Learning, to scale up the size of applicable games with already notable successes (Deepstack, Libratus, Pluribus, DeepNash). My research seeks to contribute to this endeavour by first analyzing the specifics of established methods and finding ways to introduce Hierarchical RL concepts to No-Regret learning. Please note: Updating of profile text in progress Email m.f.aichmueller@qmul.ac.uk Website LinkedIn Mastodon BlueSky GitHub Other Link Supervisor(s): Prof. Simon Lucas Dr Raluca Gaina Themes Applied Games Game AI - Previous Next
- Charlie Ringer
< Back Dr Charles Ringer University of York iGGi Alum Charlie Ringer is a researcher interested in applied Machine Learning with a focus on the ways in which we can use Deep Learning to model various facets of video games streams (e.g. stream highlights, emotional moments, in-game events, various streamer behaviours etc.). As such, his work spans many Machine Learning fields, such as Computer Vision, Affect Computing, and Natural Language Processing. His research has three motivating factors. Firstly, the challenge of how to fuse multi-view stream data (e.g. audio, web-cam footage, game footage, chat) into a single model, especially when considering the challenges of ‘in-the-wild’ data. Secondly, the untapped and bountiful data source that livestreaming represents, especially regarding the way in which streamers play games and interact with their audience. Thirdly, the exciting and emerging field of self-supervised learning which has the potential to utilise this abundance of livestream data. Charlie initially worked in the video games industry working mainly on the Magic: The Gathering - Duels of the Planeswalkers series of games before studying a BSc in Computer Science at Goldsmiths, University of London. After his BSc he joined IGGI, firstly at Goldsmiths and then at York. He was recognised as a finalist for the Twitch Research Fellowship 2019 for his research on livestream data. Email charles.ringer@york.ac.uk Website LinkedIn Mastodon BlueSky GitHub Other Link Featured Publication(s): Machine Learning with Applications From Theory to Behaviour: Towards a General Model of Engagement Modelling early user-game interactions for joint estimation of survival time and churn probability Time to die 2: Improved in-game death prediction in dota 2 Autohighlight: Highlight Detection in League of Legends Esports Broadcasts via Crowd-Sourced Data Multi-Modal Livestream Highlight Detection from Audio, Visual, and Language Data Twitchchat: A dataset for exploring livestream chat Multimodal joint emotion and game context recognition in league of legends livestreams Streaming Behaviour: Livestreaming as a Paradigm for Analysis of Emotional and Social Signals Deep unsupervised multi-view detection of video game stream highlights Streaming behaviour: Live streaming as a paradigm for multi-view analysis of emotional and social signals Rolling Horizon Co-evolution in Two-player General Video Game Playing Themes Esports Game AI Game Data - Previous Next
- Alex Fletcher
< Back Alex Fletcher Queen Mary University of London iGGi Alum Alex Fletcher is a freelance audio engineer and junior game developer working on understanding the perceived flow and player experiences in mobile rhythm games and how a dynamic difficulty adjustment system would improve these experiences. The function of EEG and other biosensors as an additional measurement of player experience is of particular interest as further research in its use as an adaptive system. Other areas of research interest include game-based learning and games with a purpose. Please note: Updating of profile text in progress Email Website LinkedIn Mastodon BlueSky GitHub Other Link Themes Applied Games Game Audio Player Research - Previous Next
- christian-guckelsberger
< Back Dr Christian Guckelsberger Queen Mary University of London iGGi Alum + Supervisor Intrinsic Motivation in Computational Creativity with Applications to Games. (Industry placement at Splash Damage and Microsoft Research) This research investigates how we can engineer artificial systems that are creative in their own right. Christian addresses this challenge with computational models of intrinsic motivation (IM). Intrinsically motivated agents perform an activity for its inherent satisfaction rather than for some instrumental outcome. A classic example is to act in order to satisfy one’s curiosity. In both theoretical and applied studies, he demonstrates that models of IM can give rise to general, robust and adaptive creative systems. Christian has shown how models of IM can be used to create highly general non-player characters. Such characters can potentially be used in a wide range of games without previous knowledge of the game mechanics, reducing costs and effort in game development while increasing robustness and behavioural variety Christian’s ongoing research stretches beyond video games, investigating the role of computational models of IM for intentional agency, open-ended development and creativity in minimal lifeforms and artificial systems. Christian studied Computer Science, History of Art and Business in Germany and the UK and is now based in London, working towards a PhD in Artificial Intelligence. His work challenges the question how computers could ever become genuinely creative with a highly interdisciplinary approach based on Computing, Cognitive Science and Philosophy. Over the last few years, he published papers on a wide range of topics, held a tutorial on intrinsic motivation in video games, organised workshops on computational serendipity and spent three months at NYU’s Game Innovation Lab for a research collaboration. Christian has substantial industry experience, looking back at three years in the R&D department of SAP SE and a recent internship at Microsoft Research Cambridge. He enjoys working in an international environment with open-minded, passionate people. Please note: Updating of profile text in progress Email Website LinkedIn Mastodon BlueSky GitHub Other Link Featured Publication(s): Not All the Same: Understanding and Informing Similarity Estimation in Tile-Based Video Games Predicting game difficulty and engagement using AI players Embodiment and computational creativity Intrinsic Motivation in Computational Creativity Applied to Videogames. PhD Thesis. 306 pages. The Relationship of Future State Maximization and von Foerster's Ethical Imperative Through the Lens of Empowerment On the Machine Condition and its Creative Expression. Understanding and Strengthening the Computational Creativity Community: A Report From The Computational Creativity Task Force. Action Selection in the Creative Systems Framework Measuring perceived challenge in digital games: Development & validation of the challenge originating from recent gameplay interaction scale (CORGIS) Generative design in Minecraft: Chronicle challenge Towards Mode Balancing of Generative Models via Diversity Weights Automating Generative Deep Learning for Artistic Purposes: Challenges and Opportunities Themes Game AI - Previous Next













