Pedro Pablo Gómez-Martín

Game-Driven Intelligent Tutoring Systems

With the increase of computer capabilities, many learning systems have become complex simulators with advanced interfaces close to game quality. However, many games features have not been added to them. This paper focus on this area, listing what games can provide to simulation-driven tutoring systems. We also describe JV2M as an example of a game-driven intelligent tutoring system to teach the compilation process of Java programs.

Query-Enabled Behavior Trees

Artificial intelligence in games is typically used for creating players opponents. Manual editing of intelligent behaviors for nonplayer characters (NPCs) of games is a cumbersome task that needs experienced designers. Our research aims to assist designers in this task. Behaviors typically use recurring patterns, so that experience and reuse are crucial aspects for behavior design. The use of hierarchical structures like hierarchical state machines, behavior trees (BTs), or hierarchical task networks, allows working on different abstraction levels reusing pieces from the more detailed levels. However, the static nature of the design process does not release the designer from the burden of completely specifying each behavior. Our approach applies case-based reasoning (CBR) techniques to retrieve and reuse stored behaviors represented as BTs. In this paper, we focus on dynamic retrieval and selection of behaviors taking into account the world state and the underlying goals. The global behavior of the NPC is dynamically built at runtime querying the CBR system. We exemplify our approach through a serious game, developed by our research group, with gameplay elements from first-person shooter (FPS) games.

Javy: Virtual Environment for Case-Based Teaching of Java Virtual Machine

Knowledge-based learning environments have become an ideal solution to provide an effective learning. Those systems base their teaching techniques upon constructivist problem solving, to supply an engaged learning environment. The students are presented with more and more challenging exercises, selected from a set of different scenarios depending on their knowledge. This paper presents a new of such systems, which aims to teach Java compilation with the help of a metaphorical virtual environment that simulates the Java Virtual Machine.

Formal Concept Analysis for Knowledge Refinement in Case Based Reasoning

Case-based Reasoning (CBR) is a problem solving paradigm that uses past experiences to solve new problems. Although CBR is supposed to alleviate the problem of knowledge acquisition, knowledge is still required to obtain the initial case base and to develop the processes of retrieving, reusing, revising and retaining cases. In this paper we propose the use of Formal Concept Analysis (FCA) to acquire and refine the knowledge available in a CBR system. In particular, we show how FC A can help to acquire indexing knowledge that supports the retrieval process, and also the use of FCA to improve the quality of the case base by identifying lack of coverage and biased combinations of case attributes.

Opportunities for CBR in learning by doing

In this paper we partially describe JV2M, a metaphorical simulation of the Java Virtual Machine where students can learn Java language compilation and reinforce object-oriented programming concepts. This description is contextualised within an abstract categorization of learning-by-doing tutoring systems intended to identify different activities where CBR can be applied. We concentrate on one of those activities, concretely on the automatic generation of new exercises through retrieval and adaptation of seed cases representing prototypical examples.

Explicit domain modelling in video games

The state-of-the-art in software engineering for game engines, recommends the use of a component-based software architecture for managing the entities in a game. A component-based architecture facilitates the definition of new types of entities as collections of components that provide basic pieces of functionality, providing a flexible software that can adapt to changes in game design. However, such flexibility comes with a price, both in terms of software understanding and error checking: a game where entity types are just run-time concepts is harder to understand than one with an explicit hierarchy of entity types; and error checking that, in a more traditional inheritance-based architecture, would come from type safety at compile time is now lost. To alleviate these problems, a component-based architecture employs blueprints, external data files that specify the particular combination of components for every entity type. In this paper we propose an extension to the component-based architecture, substituting blueprints with a full fledged domain model in OWL, including a description of the entities, its attributes and components, along with the messages they exchange. We also describe authoring tools for building such a model and show how the model improves software understanding and error checking.

Dynamic binding is the name of the game

This paper presents a tutoring system aimed at teaching how to compile Java into the language of the Java Virtual Machine, and, at the same time, promotes a better understanding of the underlying mechanisms of object-oriented programming. The interaction with the systems takes the form of a 3D videogame where the student must compete to provide the right machine instructions, collect resources needed by the instructions and use her knowledge about Java compilation to find the best strategy.

Using metaphors in game-based education

In spite of its growing popularity, due to a huge technical evolution in the last years and to the fact that new generations are more literate in games than in books, game-based teaching has been very restrictive in its application to adult education. Most applications are merely simulations built with game technology. In order to apply game-based education in domains that can not readily be turned into simulations, new ways of game design need to be explored. In this paper we describe a new approach using metaphorical worlds, and exemplify its application to teach Computer Science concepts, proposing two possible game designs for a game-based educational system devoted to teach the workings of the Java Virtual Machine.

Combining expert knowledge and learning from demonstration in real-time strategy games

Case-based planning (CBP) is usually considered a good solution to solve the knowledge acquisition problem that arises when developing AIs for real-time strategy games. Unlike more classical approaches, such as state machines or rule-based systems, CBP allows experts to train AIs directly from games recorded by expert players. Unfortunately, this simple approach has also some drawbacks, for example it is not easy to refine an existing case base to learn specific strategies when a long game session is needed to create a new trace. Furthermore, CBP may be too reactive to small changes in the game state and, at the same time, do not respond fast enough to important changes in the opponents strategy. We propose to alleviate these problems by letting experts to inject decision making knowledge into the system in the form of behavior trees, and we show promising results in some experiments using Starcraft.

Adaptation through Planning in Knowledge Intensive CBR

Adaptation is probably the most difficult task in Case-Based Reasoning (CBR) systems. Most techniques for adaptation propose ad-hoc solutions that require an effort on knowledge acquisition beyond typical CBR standards. In this paper we demonstrate the applicability of domain-independent planning techniques that exploit the knowledge already acquired in many knowledge-rich approaches to CBR. Those techniques are exemplified in a case-based training system that generates a 3D scenario from a declarative description of the training case.