The central problem that this paper addresses is how to manage dynamic change within game environ... more The central problem that this paper addresses is how to manage dynamic change within game environments in response to variable player requirements and ability. In particular, we discuss the role of a game AI to enable game systems to learn about individual user patterns, behaviours, desires or moods in order to adapt the environment in reaction to the user's interaction with the environment.
Back in the Summer of 2005, I began my focus at Linden Lab on supporting the use of Second Life f... more Back in the Summer of 2005, I began my focus at Linden Lab on supporting the use of Second Life for educational purposes. At that time there was only a handful of educators actively working in Second Life. The community consisted of about a dozen academics, with just a few projects underway on scattered plots of educational land across the grid.
Abstract We have recently developed several ways of performing Canonical Correlation Analysis [1,... more Abstract We have recently developed several ways of performing Canonical Correlation Analysis [1, 5, 7, 4] with probabilistic methods rather than the standard statistical tools. However, the computational demands of training such methods scales with the square of the number of samples, making these methods uncompetitive with eg artificial neural network methods [3, 2].
Abstract Following the lead of (Macleod 2004b) I look at the simple dice game Perudo. Presenting ... more Abstract Following the lead of (Macleod 2004b) I look at the simple dice game Perudo. Presenting an even simpler variant, I consider the probabilities of the game and present some simple heuristics for play. A simplified version of Perduo is implement and a small number of simple robot players are then built and played against each other and human players and the results presented and discussed. Finally, further research directions and extensions are considered.
Abstract. Real-Time Strategy games present an interesting problem domain for Artificial Intellige... more Abstract. Real-Time Strategy games present an interesting problem domain for Artificial Intelligence research. We review current approaches to developing AI systems for such games, noting the frequent decomposition into hierarchies similar to those found in real-world armies. We also note the rarity of any form of learning in this domain–and find limitations in the work that does use learning. Such work tends to enable learning at only one level of the AI hierarchy.
ABSTRACT There are considerable parallels between what is commonly considered to make a good gami... more ABSTRACT There are considerable parallels between what is commonly considered to make a good gaming or a good learning experience [7]. This suggests the use of computer games in the education sector. We suggest the design of a resource management computer game [4] aiming at delivering positive citizenship values.
“Second Life”, are increasingly being used in higher education institutions. As such, techniques ... more “Second Life”, are increasingly being used in higher education institutions. As such, techniques and tools are required to facilitate the delivery of learning material and assessment in these contexts. This paper introduces the early work and preliminary evaluation of such a tool, called the “quizHUD”, which has been developed alongside the SLOODLE project.
At first glance, the goal of the SLOODLE (Simulation Linked Object Oriented Dynamic Learning Envi... more At first glance, the goal of the SLOODLE (Simulation Linked Object Oriented Dynamic Learning Environment) project is to develop educational technology–specifically, software for integrating web-based virtual learning environments and 3D multi-user virtual worlds being used for educational purposes. However, a second goal is to research how such integration might best be achieved–and to understand what users might want from such technology.
These proceedings contain the papers presented at the Workshop on Adaptive approaches for Optimiz... more These proceedings contain the papers presented at the Workshop on Adaptive approaches for Optimizing Player Satisfaction in Computer and Physical Games held at the Ninth international conference on the Simulation of Adaptive Behavior (SAB'06): From Animals to Animats 9 in Rome, Italy on 1 October 2006.
Proceedings of the European Conference on …, Jan 1, 2007
Abstract: Multi-User Virtual Environments (MUVE) are seeing increasing use as educational platfor... more Abstract: Multi-User Virtual Environments (MUVE) are seeing increasing use as educational platforms, despite having a number of deficiencies for educational support–both in terms of support for student learning as well as for learning management. In comparison with web-based Virtual Learning Environments (eg Moodle, Blackboard or WebCT) we see that MUVEs have a number of advantages and some distinct weaknesses. We outline the open source Sloodle project–a project which is addressing some of these weaknesses by ...
The central problem that this paper addresses is how to manage dynamic change within game environ... more The central problem that this paper addresses is how to manage dynamic change within game environments in response to variable player requirements and ability. In particular, we discuss the role of a game AI to enable game systems to learn about individual user patterns, behaviours, desires or moods in order to adapt the environment in reaction to the user's interaction with the environment.
Back in the Summer of 2005, I began my focus at Linden Lab on supporting the use of Second Life f... more Back in the Summer of 2005, I began my focus at Linden Lab on supporting the use of Second Life for educational purposes. At that time there was only a handful of educators actively working in Second Life. The community consisted of about a dozen academics, with just a few projects underway on scattered plots of educational land across the grid.
Abstract We have recently developed several ways of performing Canonical Correlation Analysis [1,... more Abstract We have recently developed several ways of performing Canonical Correlation Analysis [1, 5, 7, 4] with probabilistic methods rather than the standard statistical tools. However, the computational demands of training such methods scales with the square of the number of samples, making these methods uncompetitive with eg artificial neural network methods [3, 2].
Abstract Following the lead of (Macleod 2004b) I look at the simple dice game Perudo. Presenting ... more Abstract Following the lead of (Macleod 2004b) I look at the simple dice game Perudo. Presenting an even simpler variant, I consider the probabilities of the game and present some simple heuristics for play. A simplified version of Perduo is implement and a small number of simple robot players are then built and played against each other and human players and the results presented and discussed. Finally, further research directions and extensions are considered.
Abstract. Real-Time Strategy games present an interesting problem domain for Artificial Intellige... more Abstract. Real-Time Strategy games present an interesting problem domain for Artificial Intelligence research. We review current approaches to developing AI systems for such games, noting the frequent decomposition into hierarchies similar to those found in real-world armies. We also note the rarity of any form of learning in this domain–and find limitations in the work that does use learning. Such work tends to enable learning at only one level of the AI hierarchy.
ABSTRACT There are considerable parallels between what is commonly considered to make a good gami... more ABSTRACT There are considerable parallels between what is commonly considered to make a good gaming or a good learning experience [7]. This suggests the use of computer games in the education sector. We suggest the design of a resource management computer game [4] aiming at delivering positive citizenship values.
“Second Life”, are increasingly being used in higher education institutions. As such, techniques ... more “Second Life”, are increasingly being used in higher education institutions. As such, techniques and tools are required to facilitate the delivery of learning material and assessment in these contexts. This paper introduces the early work and preliminary evaluation of such a tool, called the “quizHUD”, which has been developed alongside the SLOODLE project.
At first glance, the goal of the SLOODLE (Simulation Linked Object Oriented Dynamic Learning Envi... more At first glance, the goal of the SLOODLE (Simulation Linked Object Oriented Dynamic Learning Environment) project is to develop educational technology–specifically, software for integrating web-based virtual learning environments and 3D multi-user virtual worlds being used for educational purposes. However, a second goal is to research how such integration might best be achieved–and to understand what users might want from such technology.
These proceedings contain the papers presented at the Workshop on Adaptive approaches for Optimiz... more These proceedings contain the papers presented at the Workshop on Adaptive approaches for Optimizing Player Satisfaction in Computer and Physical Games held at the Ninth international conference on the Simulation of Adaptive Behavior (SAB'06): From Animals to Animats 9 in Rome, Italy on 1 October 2006.
Proceedings of the European Conference on …, Jan 1, 2007
Abstract: Multi-User Virtual Environments (MUVE) are seeing increasing use as educational platfor... more Abstract: Multi-User Virtual Environments (MUVE) are seeing increasing use as educational platforms, despite having a number of deficiencies for educational support–both in terms of support for student learning as well as for learning management. In comparison with web-based Virtual Learning Environments (eg Moodle, Blackboard or WebCT) we see that MUVEs have a number of advantages and some distinct weaknesses. We outline the open source Sloodle project–a project which is addressing some of these weaknesses by ...
Uploads
Papers by Daniel Livingstone