This chapter outlines the relation between artificial intelligence (AI) / machine learning (ML) algorithms and digital games. This relation is two-fold: on one hand, AI/ML researchers can generate large, in-the-wild datasets of human affective activity, player behaviour (i.e. actions within the game world), commercial behaviour, interaction with graphical user interface elements or messaging with other players, while games can utilise intelligent algorithms to automate testing of game levels, generate content, develop intelligent and responsive non-player characters (NPCs) or predict and respond player behaviour across a wide variety of player cultures. In this work, we discuss some of the most common and widely accepted uses of AI/ML in games and how intelligent systems can benefit from those, elaborating on estimating player experience based on expressivity and performance, and on generating proper and interesting content for a language learning game.
This paper describes the development needed to support the functional and teaching requirements of iRead, a 4-year EU-funded project which produced an award-winning serious game utilising lexical and syntactical game content. The main functional requirement was that the game should retain different profiles for each student, encapsulating both the respective language model (which language features should be taught/used in the game first, before moving on to more advanced ones) and the user model (mastery level for each feature, as reported by the student's performance in the game). In addition to this, researchers and stakeholders stated additional requirements related to learning objectives and strategies to make the game more interesting and successful; these were implemented as a set of selection rules which take into account not only the mastery level for each feature, but also respect the priorities set by teachers, helping avoid repetition of content and features, and maintaining a balance between new content and revision of already mastered features to give students the sense of progress, while also reinforcing learning.
Human activity recognition and analysis has always been one of the most active areas of pattern recognition and machine intelligence, with applications in various fields, including but not limited to exertion games, surveillance, sports analytics and healthcare. Especially in Human-Robot Interaction, human activity understanding plays a crucial role as household robotic assistants are a trend of the near future. However, state-of-the-art infrastructures that can support complex machine intelligence tasks are not always available, and may not be for the average consumer, as robotic hardware is expensive. In this paper we propose a novel action sequence encoding scheme which efficiently transforms spatio-temporal action sequences into compact representations, using Mahalanobis distance-based shape features and the Radon transform. This representation can be used as input for a lightweight convolutional neural network. Experiments show that the proposed pipeline, when based on state-of-the-art human pose estimation techniques, can provide a robust end-to-end online action recognition scheme, deployable on hardware lacking extreme computing capabilities.