Learn-and-play personalised reasoning from point-and-click to virtual reality mobile educational games

Nowadays, the use of digital games for educational purposes becomes increasingly popular. The immersive environment of a digital game causes pleasant feelings to players motivating them to participate more actively in the learning process. However, there is a variety of educational games in terms of graphics and mechanics, and each player prefers to play different game categories. For maximizing the learner’s engagement and educational results, the incorporation of the same educational content into multiple games’ categories is the solution. Nevertheless, the update of the educational content in all games increases complexity. In this paper, a mechanism that uses web services and achieves the portability of the same educational content in two completely different categories of mobile games (a point-and-click game and a virtual reality game) is presented. The gain of this innovative work is that the same educational content and mechanism can be accessible by any game’s category and platform taking advantage of the game’s characteristics, in order to motivate the student to participate in the learning process and increase her/his engagement in it. The evaluation of the presented system indicates that the integration of the same educational content to different games has a positive effect on educational results.

Integration of Video Games To Support the FCS Education Food Production Standards

Digital games can provide fun, motivating, and challenging experiences for students in the classroom. Recent research on educational games has shown that game play can positively influence academic achievement, engagement, and 21st century skills (Abdul Jabbar & Felicia, 2015; Qian & Clark, 2016). When playing games, students solve problems, collaborate with peers, and practice new skills—all within a digital environment. There are multiple games that simulate preparing and serving food. Generally, players are tasked to prepare food based on the requests of customers. Ingredients are processed on different tables or appliances, and final dishes are delivered to customers. Players who fail to complete recipes correctly or take too long during preparation may negatively influence customers' expectations. Although games vary in quality and applicability to the classroom, many address components of the Food Service and Production standard to "integrate knowledge, skills, and practices required for careers in food production and services" (National Association of State Administrators of Family and Consumer Sciences [NASAFACS], 2018, p.1).

Determining the Possibility of Transfer Learning in Deep Reinforcement Learning Using Grad-CAM (Student Abstract)

Humans are usually good at guessing whether the two games are similar to each other and easily estimate how much time to master new games based on the similarity. Although Deep Reinforcement Learning (DRL) has been successful in various domains, it takes much training time to get a successful controller for a single game. Therefore, there has been much demand for the use of transfer learning to speed up reinforcement learning across multiple tasks. If we can automatically determine the possibility of transfer learning in DRL domain before training, it could efficiently transfer knowledge across multiple games. In this work, we propose a simple testing method, Determining the Possibility of Transfer Learning (DPTL), to determine the transferability of models based on Grad-CAM visualization of the CNN layer from the source model. Experimental results on Atari games show that the transferability measure is successfully suggesting the possibility of transfer learning.

Clustering algorithm for formations in football games

In competitive team sports, players maintain a certain formation during a game to achieve effective attacks and defenses. For the quantitative game analysis and assessment of team styles, we need a general framework that can characterize such formation structures dynamically. This paper develops a clustering algorithm for formations of multiple football (soccer) games based on the Delaunay method, which defines the formation of a team as an adjacency matrix of Delaunay triangulation. We first show that heat maps of entire football games can be clustered into several average formations: “442”, “4141”, “433”, “541”, and “343”. Then, using hierarchical clustering, each average formation is further divided into more specific patterns (clusters) in which the configurations of players are different. Our method enables the visualization, quantitative comparison, and time-series analysis for formations in different time scales by focusing on transitions between clusters at each hierarchy. In particular, we can extract team styles from multiple games regarding the positional exchange of players within the formations. Applying our algorithm to the datasets comprising football games, we extract typical transition patterns of the formation for a particular team.

Evolutionary dynamics of complex multiple games

Evolutionary game theory has been successful in describing phenomena from bacterial population dynamics to the evolution of social behaviour. However, it has typically focused on a single game describing the interactions between individuals. Organisms are simultaneously involved in many intraspecies and interspecies interactions. Therefore, there is a need to move from single games to multiple games. However, these interactions in nature involve many players. Shifting from 2-player games to multiple multiplayer games yield richer dynamics closer to natural settings. Such a complete picture of multiple game dynamics (MGD), where multiple players are involved, was lacking. For multiple multiplayer games—where each game could have an arbitrary finite number of players and strategies, we provide a replicator equation for MGD having many players and strategies. We show that if the individual games involved have more than two strategies, then the combined dynamics cannot be understood by looking only at individual games. Expected dynamics from single games is no longer valid, and trajectories can possess different limiting behaviour. In the case of finite populations, we formulate and calculate an essential and useful stochastic property, fixation probability. Our results highlight that studying a set of interactions defined by a single game can be misleading if we do not take the broader setting of the interactions into account. Through our results and analysis, we thus discuss and advocate the development of evolutionary game(s) theory, which will help us disentangle the complexity of multiple interactions.

Cyclic dominance emerges from the evolution of two inter-linked cooperative behaviours in the social amoeba

Evolution of cooperation has been one of the most important problems in sociobiology, and many researchers have revealed mechanisms that can facilitate the evolution of cooperation. However, most studies deal only with one cooperative behaviour, even though some organisms perform two or more cooperative behaviours. The social amoeba Dictyostelium discoideum performs two cooperative behaviours in starvation: fruiting body formation and macrocyst formation. Here, we constructed a model that couples these two behaviours, and we found that the two behaviours are maintained because of the emergence of cyclic dominance, although cooperation cannot evolve if only either of the two behaviours is performed. The common chemoattractant cyclic adenosine 3′,5′-monophosphate (cAMP) is used in both fruiting body formation and macrocyst formation, providing a biological context for this coupling. Cyclic dominance emerges regardless of the existence of mating types or spatial structure in the model. In addition, cooperation can re-emerge in the population even after it goes extinct. These results indicate that the two cooperative behaviours of the social amoeba are maintained because of the common chemical signal that underlies both fruiting body formation and macrocyst formation. We demonstrate the importance of coupling multiple games when the underlying behaviours are associated with one another.