Research Publications

The authors study a variant of Cops and Robbers in which robbers aim to maximize the number of damaged vertices. They verify a conjecture about the vertices a single cop can protect on triangle-free graphs and disprove it for general graphs. They also initiate the study of two cops against two robbers, determining exact damage numbers on cycles and providing bounds on paths.

This qualitative case study explores a Game Friendly School project in Türkiye through responses from five first-grade teachers and drawings and short narratives from 98 students. Participants described enjoyment, friendship, collaboration and a positive school atmosphere, while teachers raised age-appropriateness and safety considerations. The findings describe perceived experiences rather than measuring the project’s effectiveness.

This paper examines whether modern rules proposed for the ancient game Pente Grammai work as playable game reconstructions.

The authors develop counterparts of infinitesimals and atomic weight theory for dicotic games under the misère convention. They introduce muptimals and identify tame rulesets where normal-play atomic weight methods remain useful up to specified endgames. They also demonstrate limits of this approach outside dicotic games: full misère play has no infinitesimal elements and does not satisfy the Archimedean property.

This qualitative study follows 28 pre-service mathematics teachers as they use games to develop mathematical generalizations. It examines how gameplay, representations and discussion support pattern recognition and symbolic reasoning, and where participants need further support for justification and proof.

This paper studies the smallest graphs of a given minimum degree on which Dominator loses when moving first in the Maker Breaker Domination Game. It determines exact values for minimum degrees up to three and improves the general lower bound.

An object found in Heerlen, the Netherlands was interpreted as a Roman board game, but the pattern of lines on its surface do not reflect that of any known game. Use-wear analysis shows that its surface is smoothed in a manner consistent with abrasion by Roman game pieces. Simulated play with artificial intelligence (AI) placed rules from Northern European games on six possible board geometries based on the pattern on the object’s surface. These simulations identified rules that produced disproportionate movement along the worn lines, supporting a conclusion that the object may have been used to play a blocking game.

This study examines stone-carved game boards in the theatre at Aphrodisias. Their locations in the seating area and on the stage suggest that gaming supported the theatre’s role as a social space beyond performances.

This paper studies how chess AI can reproduce the individual playing styles of 14 elite players. It combines a Mixture of Experts with Low-Rank Adaptation (MoE-LoRA) and limited Monte Carlo Tree Search to improve behavioural alignment while reducing repetition and tactical errors. Evaluation on recurring board positions measures both move accuracy and how closely the model follows each player’s observed choices.

This chapter examines alquerque game boards carved into the fabric of León Cathedral as evidence of medieval play and the social use of religious space. Their spatial distribution and archaeological contexts offer insight into everyday practices and the transmission of gaming traditions. The contribution acknowledges GameTable COST Action CA22145.

This contribution concerns real-time procedural generation of coherent, infinite two-dimensional landscapes containing multiple biomes.

This ethnographic study examines variations in the rules of nərd, dama and mancala-style games in Azerbaijan. It explores how player communities, cultural transmission and the material form of game boards influence the way rules are maintained or adapted.

This article examines Maurice Kraïtchik’s contributions to recreational mathematics.

This action research study examines how tabletop games can support teacher education. Master’s students at the University of Tirana explored traditional and modern games, adapted them to learning objectives, and tested them through peer microteaching. The findings suggest benefits for engagement, collaboration, conceptual understanding and reflective teaching practice, connecting the educational aims of GameTable with classroom application.

This report presents the outcomes of the GameTable meeting held in Prague on 19–20 February 2026 to open the third grant period. It covers collaborative projects connecting cultural heritage and game AI, human-like AI research, and future opportunities for collaboration beyond the Action.

Valet provides a common benchmark of 21 traditional imperfect-information card games spanning different cultures, genres and game structures. Their rules are encoded in the RECYCLE card game description language. The paper characterizes the games through simulations and establishes baseline results for Monte Carlo Tree Search against random opponents.

We present an efficient and generalised procedure to accurately identify the best (or near best) performing algorithm for each sub-task in a multi-problem domain. Our approach treats this as a set of best arm identification problems for multi-armed bandits, where each bandit corresponds to a specific task and each arm corresponds to a specific algorithm or agent. We propose an optimistic selection process based on a chosen confidence interval, that ranks each arm across all bandits in terms of their potential to influence our overall simple regret. We evaluate the performance of our approach on two of the most popular general game playing domains, the General Video Game AI (GVGAI) framework and the Ludii general game playing system, with the goal of selecting a high-performing agent for each game using a limited number of available trials. Compared to previous best arm identification algorithms for multi-armed bandits, our results demonstrate a substantial performance improvement in terms of average simple regret and average probability of error. This novel approach can be used to significantly improve the quality and accuracy of agent evaluation procedures for general game frameworks, as well as other multi-task domains with high algorithm runtimes.

Les jeux constituent une source essentielle pour comprendre les sociétés antiques, mais leur étude reste souvent fragmentaire. Le projet Graffiti in the shadow of the Acropolis, mené à l’Université de Lausanne, analyse les graffiti non textuels d’Attique — souvent interprétés comme des plateaux de jeu — en les étudiant dans leur contexte architectural et social. Grâce à une combinaison de méthodes archéologiques, de technologies numériques (SIG, photogrammétrie, 3D) et d’outils computationnels, il renouvelle l’interprétation de ces marques et met en lumière le rôle du jeu comme pratique sociale inscrite dans l’espace urbain, révélant des usages quotidiens longtemps invisibles.

This forthcoming chapter examines Roman gaming equipment from Iža (Celemantia). Publication details and a PDF will be added when available.

The LAMIR algorithm learns an abstract model of an imperfect-information game through interaction. It uses that model for look-ahead reasoning at test time, improving trained agents while keeping the search manageable in large games.

This study uses artificial intelligence to reconstruct Coptic game boards in three dimensions from a single image. Its open-source workflow combines image preprocessing, mesh generation and geometric optimization, supporting the digital preservation and study of inaccessible archaeological objects.

This research explores the development of an advanced AI-driven dialogue simulator specifically designed to assist educators in effectively practicing and navigating various pedagogical scenarios through role-playing. The system integrates state-of-the-art generative AI with virtual reality (VR), utilizing dynamic 3D avatars capable of interactive dialogue and real-time question generation based on predefined knowledge and user input. Educators can engage in realistic simulations, receiving personalized and contextually relevant questions and feedback from AI-generated virtual interlocutors. By leveraging speech-to-text technology, user interactions are seamlessly processed to generate immediate, targeted responses, facilitating an immersive and responsive learning environment. This dialogue simulator aims to enhance educators’ preparedness, adaptability, and communication skills by allowing iterative practice in a safe virtual setting. Technical implementations include the utilization of open-source large language models, real-time avatar lip synchronization, customizable virtual personas, and sophisticated scenario generation methods. Initial evaluations were conducted to assess the feasibility and usability of the VR-based dialogue simulator. Specifically, exploring whether educators found the system immersive, realistic, and helpful for practicing pedagogical scenarios. Future research will further refine AI responsiveness and expand scenario complexity, contributing meaningfully to AI-assisted pedagogical training methodologies.

This contribution examines mathematical games presented in the journal Sphinx between 1931 and 1939.

This paper reviews findings from cognitive psychology and previous approaches to human-like artificial agents, considering their application to General Game Playing. It introduces CogniPlay, a model under development that aims to reflect human decision-making through chunk-based representations, dual-process cognition and selective search.

Archaeological finds connected to gaming and play are seldom published in detail from Roman Pannonia, although these are frequently found during excavations. Therefore, the area of Pannonia along the Danube limes represents a blank spot in the research of the ludic culture of the Roman Empire. The paper presents a comprehensive study of the pieces of Roman gaming equipment in the collection of the Kuny Domokos Museum in Tata (Hungary) and discusses sixty hitherto unpublished gaming paraphernalia including six-sided cubic bone dice, bone, glass and ceramic counters, as well as ceramic marbles. As almost all objects were found in the settlement complex of Brigetio (Komárom/Szőny, Hungary), the close examination of the artefacts lets us gain insight into the gaming practices of Roman Brigetio.

This article examines a Roman period gameboard (Gameboard 1) from Gerulata (modern Rusovce, Slovakia), a military settlement on the Danube frontier in Pannonia. The board was carved into a tile and features a 7x8 grid layout. A much smaller fragment of another possible gameboard or fragment of wall tile (Possible gameboard/tile 2) was also recovered from the site, though this identification is less secure. This study explores the archaeological context, typology, and potential identification of the board(s). Gameboard 1 was probably used for playing the game known as ludus latrunculorum, even if other possibilities should not be entirely discarded. Analysis of published and unpublished finds from the Northern Carpathian basin and neighbouring regions—which we present together for the first time—indicates that this gameboard belongs to a wider ludic culture, which was particularly visible among Roman soldiers. Portable elements of Roman-style material culture (dice, counters) but not bulky gameboards have been recovered in parts of Slovakia that were in the Barbaricum, and this suggests some cultural diffusion across borders in the Roman period. These objects appear to be prestige finds, since they are commonly associated with elite sites, but we cannot be sure that they were used to play ‘Roman’ games. The article argues that such objects offer a valuable lens into the social and cultural life of Roman soldiers and civilians on the frontier. It also highlights the methodological challenges of identifying and interpreting gameboards in provincial contexts.

This study presents the design and development of a 2D narrative game set in a 3D environment, leveraging Artificial Intelligence (AI) to deliver interactive age-tailored and educational historical storytelling. The game combines 2D-style narratives with 3D settings to create an immersive historical experience. Players interact with a companion character powered by Large Language Model Meta AI (Llama), an advanced language model, at key checkpoints. The AI delivers historically accurate content, adapting its storytelling to different age groups (children, teens, and adults) to ensure accessibility and sustained engagement. The game promotes interactive learning by integrating gameplay with period-specific historical narratives. To ensure accuracy, Llama is configured to respond strictly within the context of each historical period. Players can engage in dialogue with the AI for context-appropriate guidance, supported by a Python-based API that dynamically adjusts narratives based on player progress and age. This study explores the integration of Natural Language Processing (NLP) in games to enhance educational experiences, showcasing how age-adaptive storytelling and interactive dialogue improve historical learning through gameplay. The study details the game’s architecture and its balance of academic content with immersive gameplay. This innovative approach highlights AI’s transformative potential in digital learning.

This paper presents Generalized Proof-Number Monte- Carlo Tree Search: a generalization of recently proposed combinations of Proof-Number Search (PNS) with Monte-Carlo Tree Search (MCTS), which use (dis)proof numbers to bias UCB1-based Selection strategies towards parts of the search that are expected to be easily (dis)proven. We propose three core modifications of prior combinations of PNS with MCTS. First, we track proof numbers per player. This reduces code complexity in the sense that we no longer need disproof numbers, and generalizes the technique to be applicable to games with more than two players. Second, we propose and extensively evaluate different methods of using proof numbers to bias the selection strategy, achieving strong performance with strategies that are simpler to implement and compute. Third, we merge our technique with Score Bounded MCTS, enabling the algorithm to prove and leverage upper and lower bounds on scores—as opposed to only proving wins or not-wins. Experiments demonstrate substantial performance increases, reaching the range of 80% for 8 out of the 11 tested board games.

This article explores the role of Working Group 5 (WG5) within the GameTable COST Action (CA22145), a pan- European initiative dedicated to the computational study and public dissemination of traditional games as cultural heritage. WG5 focuses on the educational and public engagement dimensions of this mission, translating research on historical games into pedagogical tools, participatory experiences, and digital applications. Organized into three focus groups, Education, Digital Exhibition, and the General Game System, WG5 has developed a range of outputs, including teacher booklets, glossaries, an artificial intelligence-assisted adventure game for educators, and a citizen science-enabled digital exhibition. It has also proposed a pan-European school tournament hosted on the Ludii platform, designed to promote game literacy and intercultural learning. Finally, WG5 contributes to GameTable’s sustainability goals through plans for resource hubs, continued training programs, and museum partnerships. The article situates these activities within broader frameworks of game-based learning, gamification, and participatory digital heritage, demonstrating how games can serve as powerful tools for inclusive education, heritage interpretation, and cross-sector collaboration.

Recent advances in board game agents have led to superhuman performances in various games, but their decision-making processes remain opaque even to experienced players and are not cognitive-plausible from a human perspective. Based on criteria for cognitive-plausible systems, we introduce an approach that emphasizes interpretable knowledge representation through pattern-based and hierarchical structures. Our approach uses Genetic Programming to learn feature programs that capture spatial relationships on the game board. In addition, it utilizes decision trees to model agent behavior based on these feature programs. The result is a hierarchical, pattern-based model for generating post-hoc explanations for board game agents. We show that our method generates comprehensible explanations in Tic-Tac-Toe. We then discuss the current limitations in generalizing the approach to more complex games using Connect 4 as an example.

Large language models (LLMs) have shown impressive capabilities in generating program code, opening exciting opportunities for applying program synthesis to games. In this work, we explore the potential of LLMs to directly synthesize usable code for a wide range of gaming applications, focusing on two programming languages, Python and Java. We use an evolutionary hill-climbing algorithm, where the mutations and seeds of the initial programs are controlled by LLMs. For Python, the framework covers various game-related tasks, including five miniature versions of Atari games, ten levels of Baba is You, an environment inspired by Asteroids, and a maze generation task. For Java, the framework contains 12 games from the TAG tabletop games framework. Across 29 tasks, we evaluated 12 language models for Python and 8 for Java. Our findings suggest that the performance of LLMs depends more on the task than on model size. While larger models generate more executable programs, these do not always result in higher-quality solutions but are much more expensive. No model has a clear advantage, although on any specific task, one model may be better. Trying many models on a problem and using the best results across them is more reliable than using just one.

We examine how the shift from the Royal Game of Ur to the “game of twenty squares” altered play experience. Using Irving Finkel’s rules—grounded in a 177 BCE Seleucid tablet and ethnographic analogies—we run AI self-play on both boards to quantify gameplay differences. Metrics include game length (turns), first-player advantage, and “drama” (turn-by-turn win-probability swings). We relate these results to the games’ changing geography and chronology, suggesting that the board’s geometric redesign produced systematically different dynamics and may reflect evolving gameplay preferences in the ancient Near East.

In imperfect-information games, agents must make decisions based on partial knowledge of the game state. The Belief Stochastic Game model addresses this challenge by delegating state estimation to the game model itself. This allows agents to operate on externally provided belief states, thereby reducing the need for game-specific inference logic. This paper investigates two approaches to represent beliefs in games with hidden piece identities: a constraint-based model using Constraint Satisfaction Problems and a probabilistic extension using Belief Propagation to estimate marginal probabilities. We evaluated the impact of both representations using general-purpose agents across two different games. Our findings indicate that constraint-based beliefs yield results comparable to those of probabilistic inference, with minimal differences in agent performance. This suggests that constraint-based belief states alone may suffice for effective decision-making in many settings.

The multifaceted study of games spans diverse disciplines which have often operated in isolation. In this era of rapid developments in computer science, games stand as testbeds for pioneering methodologies, shaping and influencing techniques in economics, engineering, and AI. At the same time, archaeologists, historians, and anthropologists explore the motivations underlying human play, dissecting its social implications on both individual and societal levels. Furthermore, games have played a central role in pedagogical development and are progressively acknowledged as integral components of humanity’s cultural heritage.
The GameTable COST Action is an ongoing scientific network which started in October 2023, funded by the European Cooperation in Science and Technology research scheme, aiming to gather scholars, researchers and stakeholders from a variety of fields and career stages across academia, industry and heritage institutions. GameTable aims to forge long-lasting international collaborations by establishing an interdisciplinary network to develop new methodologies and applications for using game AI to study, reconstruct, and preserve the intangible cultural heritage of games.
The network proposes to approach AI techniques, archaeology, and gaming research into a cohesive framework. In doing so, the group aspires to demonstrate how games are both form of entertainment and repositories of cultural richness and historical narratives. Key challenges include innovative approaches for studying and preserving heritage games, reconstructing missing rules in incomplete games, simulating human-like play, applying AI to study historical games, developing novel methods to identify unrecognized games measuring their evolution across space and time, developing pedagogical tools, and leveraging AI in a culturally diverse way.

On January 20 and 21, 2025, the GameTable COST Action’s Working Group 1 convened in London for a meeting focused on game-playing AI, including search algorithms, knowledge representation, and reinforcement learning. The first day featured participant presentations, while the second day was dedicated to plenary and small group discussions. A significant outcome was the identified need for accessible resources to bridge AI and cultural heritage research. This report provides a summary of the discussions and talks that took place during the meeting.

The rapid advancement of artificial intelligence (AI) in game studies has opened new avenues for analyzing and categorizing traditional games. This research proposes the development of a comprehensive, structured dataset of game board photographs to train supervised learning models capable of identifying and ranking potential matches from the extensive Ludii game database . The dataset will consist of images contributed by GameTable network members and partners, capturing diverse game boards across various cultural and historical contexts. Each image will be expertly annotated to serve as input for supervised learning algorithms, ensuring a robust training set that encompasses a wide variety of games and board configurations. Machine learning models, such as convolutional neural networks (CNNs), will be employed to recognize patterns in the game board images and rank potential matches from the Ludii database. The ultimate goal is to develop a user-friendly application, accessible via web and Android platforms, enabling users to upload photos of board games and receive ranked suggestions on potential matches. This tool aims to assist professional archaeologists in identifying and contextualizing gaming artifacts during excavations, thereby enhancing on-site research and interpretation. By integrating AI-driven image recognition with traditional game studies, this project seeks to democratize access to game knowledge, support game preservation, and contribute to the broader field of AI applications in cultural heritage.

As commercial tabletop and video games become increasingly popular throughout the world, traditional board games are becoming a form of endangered intangible cultural heritage, and in many places what were once widely played games have largely been abandoned. The Digital Ludeme Project applies artificial intelligence and computational techniques to the incompletely documented rules of traditional board games from 3500 BCE until the present to help preserve accurate knowledge of these games. The project engages with the historiography of knowledge about traditional games, determining which parts of the rules are provided by various textual, archaeological, artistic, and ethnographic sources. To document this knowledge and situate it geographically and chronologically, the Ludii Games Database provides the first comprehensive and rigorous Open Access database of scholarship on the global history of board games. Furthermore, the database is used as a source for data for the Ludii general game system, which makes the games playable and provides the analytical tools that will be used in future reconstruction work. Finally, the database may also serve as a reference tool for scholars pursuing research on games-related topics.

Games, as an essential form of intangible cultural heritage, have played a special role in the intellectual development of people. Various forms of games have emerged throughout history and spread in Azerbaijan. This article focuses mainly on board games, discussing their history and current rules of play. It also provides comparisons with similar types of games in other countries and explores the migration of game rules.

The field of General Game Playing (GGP) usually yearns to create agents that are able to win any game as efficiently as possible. Utility functions are then easy to find: winning the game, or achieving a better score, nets you better results. This resulted in game playing becoming a testbed for artificial intelligence. New techniques are often tested on games such as Chess, Go or Checkers where the environments are easy to formally define, while the task is still being tied closely to human intelligence. Superhuman-level has been achieved in a great number of games past the 2010s. The first notable instance was Chess world champion Garry Kasparov losing to DeepBlue, a chess playing computer, in 1997 during a six-game match. Researchers became interested in games where heuristics were harder to encode, such as Go. Only in 2016 did AlphaGo become the first computer Go program to beat a 9-dan professional player, Lee Sedol. It did so using novel techniques, such as Deep Neural Networks (DNN) trained using self-play to evaluate positions and Monte-Carlo Tree Search (MCTS) to search the game tree, which are now staples of superhuman-level game playing agents. Such level of play was only achieved in agents playing specific games. In such instances, an optimized game representation and evaluation functions can be built by leveraging domain-specific knowledge. General Game Playing (GGP) aims to generalize game playing agents so that they can pick up any game. In this more general setting, domain-specific knowledge can only be used scarcely, leading to more polyvalent yet weaker a...

Automatically generating novel and interesting games is a complex task. Challenges include representing game rules in a computationally workable form, searching through the large space of potential games under most such representations, and accurately evaluating the originality and quality of previously unseen games. Prior work in automated game generation has largely focused on relatively restricted rule representations and relied on domain-specific heuristics. In this work, we explore the generation of novel games in the comparatively expansive Ludii game description language, which encodes the rules of over 1000 board games in a variety of styles and modes of play. We draw inspiration from recent advances in large language models and evolutionary computation in order to train a model that intelligently mutates and recombines games and mechanics expressed as code. We demonstrate both quantitatively and qualitatively that our approach is capable of generating new and interesting games, including in regions of the potential rules space not covered by existing games in the Ludii dataset.

Imperfect-information games present significant challenges for General Game Playing (GGP) agents. Traditional models have limitations that hinder their applicability in this domain. These models require agents to construct and maintain estimates about the game state, a process that is often game-specific and can unintentionally introduce domain-specific knowledge. Furthermore, this specificity undermines the core goal of GGP to generate domain-independent strategies. To overcome these challenges, we propose the Belief Stochastic Game model. This novel framework shifts the responsibility of state estimation from the agent to the game model itself, allowing agents to focus solely on strategy development. This externalisation of the state estimation process is enabled by the exploitation of the common structures found in many imperfect-information games. The new model facilitates the development of more general agents that can adapt to a wide range of games.

Étienne de Flacourt's Histoire de la grande isle Madagascar is renowned for its detailed descriptions of Madagascar's flora, fauna, and local culture. The posthumously published second edition (1661) includes the earliest known record of a four-row mancala game, accompanied by a drawing and a detailed explanation using local terminology. These rules, resembling the modern game of Bao (known as Katra-be in Madagascar), suggest a long history of four-row mancala games, despite a lack of archaeological evidence. As governor of a French trading post in Fort Dauphin, de Flacourt's work endures, offering a glimpse into 17th-century Madagascar.

Games, as a subject of study, span diverse disciplines, often in isolation. In computer science and mathematics, games stand as testbeds for pioneering methodologies, shaping state-of-the-art techniques in economics, engineering, and AI. Simultaneously, archaeologists, historians, and anthropologists explore the motivations underlying human play, dissecting its social implications on both individual and societal levels. Games have not only been instrumental in pedagogical development but are gaining recognition as integral components of humanity's intangible cultural heritage. However, this recognition is tinged with a poignant realization: much of the world's game heritage has been lost, overshadowed by colonialism, imperialism, and commercialization. Against this backdrop, the paper introduces the GameTable Network, an EU-funded COST Action overcoming disciplinary divides. The network's research integrates AI techniques, archaeology, and gaming research, into a cohesive framework. In doing so, the network aspires to rekindle the essence of games not only as mere entertainment but also as repositories of cultural richness and historical narratives. The aim is not only to uncover the mysteries of the past but to acknowledge and rectify the gaps hindering a holistic understanding of games. Our objectives span from reconstructing historical games to advancing AI methodologies, from fostering educational programs to embracing community involvement in digital initiatives. Key challenges include innovative approaches for studying, reconstructing, and preserving heritage games, reconstructing missing rules in incomplete games, simulating play at a human level, applying AI to study historical games, developing novel methods to identify unrecognized games measuring their evolution across space and time, developing pedagogical tools, and leveraging GameAI in a culturally diverse way, all crucially contributing to the study of heritage games. Through this interdisciplinary lens, our aspiration is to enrich scholarly dialogue within the Eastern Mediterranean, Europe and beyond, advancing our collective understanding of the historical and cultural dimensions embedded in games through AI-driven research.

The inaugural in-person meeting for the "GameTable" COST Action’s Working Group 1(WG1) on Search, Planning, Learning,and Explainability took place at the Leiden Institute of Advanced Computer Science (LIACS) on January 31st,2024. The primary aims of this meeting were to facilitate talks and discussions on, and connect researchers interested in, three core research goals: (1) human-like game-playing AI, (2) imperfect-information games within a general game playing context, and (3) explainable search and reinforcement learning in games. This report provides a summary of the discussions and talks that took place during the meeting.

Games with imperfect information represent a major challenge for General Game Playing (GGP) agents. To address this, we propose to extend Ludii, a state-of-the-art GGP system, by improving the game description to include hidden information, adopting a formalism and modelling closer to reinforcement learning, and adapting the search algorithms to make better use of the game description. These improvements will make it possible to simulate and analyse a wide range of games with imperfect information.

The GameTable COST Action kickoff, focusing on “Computational Techniques for Tabletop Games Heritage,” took place at Leiden University in the Pieter de la Court Building from January 29th to 30th, 2024. This event aimed to convene researchers from diverse backgrounds involved in the Action, offering an opportunity to present an overview of the key research areas, share concrete case studies, and facilitate discussions and idea exchanges across fields that may not typically intersect, thereby enhancing the organization of the Action. This report provides a summary of the organization and discussions of the event, and future plans for GameTable.