Matthew Stephenson

ORCID: 0000-0002-3867-5842
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Artificial Intelligence in Games
  • Digital Games and Media
  • Reinforcement Learning in Robotics
  • Educational Games and Gamification
  • Sports Analytics and Performance
  • Video Analysis and Summarization
  • Human Motion and Animation
  • Mathematics and Applications
  • AI-based Problem Solving and Planning
  • Computer Graphics and Visualization Techniques
  • Evolutionary Algorithms and Applications
  • Topic Modeling
  • Anomaly Detection Techniques and Applications
  • Multi-Agent Systems and Negotiation
  • Music and Audio Processing
  • Gambling Behavior and Treatments
  • Teaching and Learning Programming
  • Crime Patterns and Interventions
  • Advanced Malware Detection Techniques
  • Optimal Experimental Design Methods
  • Image Processing and 3D Reconstruction
  • Multimodal Machine Learning Applications
  • Natural Language Processing Techniques
  • Robotic Path Planning Algorithms
  • Research Data Management Practices

Flinders University
2015-2025

Maastricht University
2018-2024

University of Central Florida
2018-2022

Friedrich-Alexander-Universität Erlangen-Nürnberg
2022

Kyoto University
2022

Universidad Pontificia Comillas
2022

Emerald Group Publishing (United Kingdom)
2022

Australian National University
2016-2021

The Behavioural Insights Team
2021

University of New Brunswick
2021

10.1038/s42256-019-0072-x article EN Nature Machine Intelligence 2019-07-09

While current General Game Playing (GGP) systems facilitate useful research in Artificial Intelligence (AI) for game-playing, they are often somewhat specialised and computationally inefficient. In this paper, we describe the "ludemic" general game system Ludii, which has potential to provide an efficient tool AI researchers as well designers, historians, educators practitioners related fields. Ludii defines games structures of ludemes -- high-level, easily understandable concepts allows...

10.48550/arxiv.1905.05013 preprint EN other-oa arXiv (Cornell University) 2019-01-01

This paper presents a procedural generation algorithm for levels in physics-based puzzle games similar to Angry Birds. The proposed creates consisting of various self-contained structures placed throughout 2D area. Each structure can be either on the ground or atop floating platforms within available level space. These are created using variety different block types and do not require predefined substructures composite elements. Target object locations determined based combination factors,...

10.1609/aiide.v12i1.12871 article EN Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment 2021-06-25

This paper presents a procedural level generation algorithm for physics-based puzzle games similar to Angry Birds. The proposed is capable of creating varied, stable and solvable levels consisting multiple self-contained structures placed throughout 2D area. work presented in this builds improves upon previous algorithm, enhancing it several ways. created are evaluated based on updated fitness function which considers key structural aspects, including both robustness variety. results...

10.1109/cig.2017.8080448 article EN 2017-08-01

This paper presents an overview of the second AIBIRDS level generation competition, held jointly at 2017 IEEE Conference on Computational Intelligence and Games 26th International Joint Artificial Intelligence. competition tasked entrants with developing a generator for physics-based puzzle game Angry Birds. Submitted generators were required to deal many physical reasoning constraints caused by realistic nature game's environment, in addition ensuring that created levels fun, challenging,...

10.1109/tg.2018.2854896 article EN IEEE Transactions on Games 2018-07-12

Research Summary Firms exhibit heterogeneity in size, productivity, and internal structure, this is true even within the same industry. Our paper provides evidence of a link between an organization's culture—specifically trust environment—and its level specialization. We show experimentally that exogenously‐imposed culture endogenously leads to variation organizational form. prime demonstrate organization affects division labor consequently productivity nonhierarchical teams. This consistent...

10.1002/smj.3024 article EN Strategic Management Journal 2019-03-20

This paper presents a procedural content generation algorithm for the physics-based puzzle game Angry Birds. The proposed creates complex stable structures using variety of 2D objects. These are generated without aid pre-defined substructures or composite elements. created evaluated based on fitness function which considers several important structural aspects. results this analysis in turn affects likelihood particular objects being chosen future generations. Experiments were conducted...

10.1109/cig.2016.7860410 article EN 2016-09-01

Over the past few years Angry Birds AI competition has been held in an attempt to develop intelligent agents that can successfully and efficiently solve levels for video game Birds. Many different strategies have developed complex challenging physical reasoning problems associated with such a game. However, performance of these various is non-transitive varies significantly across levels. No single agent dominates all situations presented, indicating procedures are better at solving certain...

10.1609/aiide.v13i1.12946 article EN Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment 2021-06-25

Many games often share common ideas or aspects between them, such as their rules, controls, playing area. However, in the context of General Game Playing (GGP) for board games, this area remains under-explored. We propose to formalise notion "game concept", inspired by terms generally used game players and designers. Through Ludii System, we describe concepts several levels abstraction, itself, moves played, states reached. This new GGP feature associated with ludeme representation opens...

10.1109/cog52621.2021.9618990 article EN 2021 IEEE Conference on Games (CoG) 2021-08-17

Ludii is a general game system being developed as part of the ERC-funded Digital Ludeme Project (DLP). While its primary aim to model, play, and analyse full range traditional strategy games, also has potential support wide AI research topics competitions. This paper describes some future competitions challenges that we intend run using system, highlighting most important aspects can potentially lead many algorithm improvements new avenues research. We compare contrast our proposed...

10.1109/cig.2019.8848084 article EN 2021 IEEE Conference on Games (CoG) 2019-08-01

In its most important administrative law decision in decades, the Supreme Court West Virginia v. Environmental Protection Agency embraced an aggressive version of so-called “major questions doctrine” (MQD), which appears to require unusually explicit statutory authorization before agencies may undertake “major” regulatory actions. The majority claimed that this strong MQD is based on longstanding precedent and will have salutary effects policymaking process. This Article argues neither claim...

10.2139/ssrn.4409630 article EN SSRN Electronic Journal 2023-01-01

The General Video Game AI Framework has featured multiple games and several tracks since the first competition in 2014. Although of framework are very assorted nature, there is an underlying commonality with respect to physics that govern game: all them based on a grid where sprites make discrete movements, which not expressive enough cover any meaningful physics. This paper introduces enhanced system brings real-world such as friction, inertia other forces framework. We also introduce...

10.1109/cig.2017.8080443 article EN 2017-08-01

10.1016/j.artint.2019.103232 article EN publisher-specific-oa Artificial Intelligence 2020-01-06

This paper investigates the performance of different general-game-playing heuristics for games in Ludii general game system. Based on these results, we train several regression learning models to predict based each game's description file. We also provide a condensed analysis available Ludii, and ludemes that define them.

10.1109/cog52621.2021.9619052 article EN 2021 IEEE Conference on Games (CoG) 2021-08-17

Although researchers and the public intuitively know that lack of available data on topic officer-involved shootings (OIS) is a problem, scope problem has not been identified. A transparency regarding OIS contributes to legitimacy crisis facing policing in United States. Valid reliable would create ability craft assess informed policy. In addition, police organizations' roles could then be evaluated allowing for accountability additional training based incidents. Employing content analysis...

10.1177/0886260520913646 article EN Journal of Interpersonal Violence 2020-03-31

Due to the emergence of AI systems that interact with physical environment, there is an increased interest in incorporating reasoning capabilities into those systems. But it enough only have operate a real environment? In world, we constantly face novel situations not encountered before. As humans, are competent at successfully adapting situations. Similarly, agent needs ability function under impact novelties order properly open-world environment. To facilitate development such systems,...

10.1016/j.artint.2024.104198 article EN cc-by-nc Artificial Intelligence 2024-08-02

Over the past few years Angry Birds AI competition has been held in an attempt to develop intelligent agents that can successfully and efficiently solve levels for video game Birds. Many different strategies have proposed complex challenging physical reasoning problems associated with such a game. The performance of these increased significantly over competition's lifetime thanks approaches improved techniques employed. However, there still exist key flaws within designs often lead them make...

10.1145/3235765.3235775 article EN 2018-08-07

The Digital Ludeme Project (DLP) aims to reconstruct and analyse over 1000 traditional strategy games using modern techniques. One of the key aspects this project is development Ludii, a general game system that will be able model play complete range required by project. Such an undertaking create wide possibilities for new AI challenges. In paper we describe many features Ludii can used. This includes designing modifying description language, creating agents capable playing these games,...

10.1109/cig.2019.8847949 article EN 2021 IEEE Conference on Games (CoG) 2019-08-01

This paper presents a structure generation algorithm, which converts rough human drawings into stable structures comprising rectangular blocks, suitable for physics-based 2-D environments. Generating viable environment imposes many additional requirements above those of most traditional sketch-based domains. Our method is sophisticated enough to deal with these requirements, while still ensuring that the generated accurately represents original sketch. We describe and implement framework...

10.1109/tg.2019.2908353 article EN IEEE Transactions on Games 2019-04-01
Coming Soon ...