Multi-agent reinforcement learning for character control

Simultaneous control of multiple characters has been a research topic that has been extensively pursued for applications in computer games and computer animations, for applications such as crowd simulation, controlling two characters carrying objects or fighting with one another and controlling a team of characters playing collective sports. With the advance in deep learning and reinforcement learning, there is a growing interest in applying multi-agent reinforcement learning for intelligently controlling the characters to produce realistic movements. In this paper we will survey the state-of-the-art MARL techniques that are applicable for character control. We will then survey papers that make use of MARL for multi-character control and then discuss about the possible future directions of research.

Development of Virtual Laboratory Simulation: e-SCILAB on Waves for Grade 7 Science

Virtual Laboratory Simulation (VLS) enhances student understanding, retention of concepts, and promotes active learning through direct visualization and manipulation of computer animations. This study aimed to develop a VLS with complementary manual for Grade 7 physics that can be utilized by the teachers to address the needs of the students in the new normal learning. The developed VLS is composed of a Teacher’s Instructional Manual and Student’s Learning Manual. The study used Research and Development approach with the ADD (Analysis, Design, and Development) model as the instructional method. These allows the researchers to analyze the need of VLS, identify the design of VLS, and develop the VLS. The VLS was evaluated based on the content quality with a mean of 4.68, technical quality with 4.65, and instructional quality with 4.57. The results for the complementary activity manual in terms of the content quality is a mean of 4.62, the technical quality with 4.54, and the instructional quality with 4.55. The result showed that the developed VLS together with the complementary activity manual satisfied the quality components and attained the intended standards. It is recommended that evaluation of the effectiveness of VLS through pilot tests in public and private schools.

Basic reflections on the implementation of different 3D technologies co-operating in the architectural design process

The constant progress of technique is inevitable nowadays and seems to be the same in a predictable future. The observation of this phenomenon leads us to formulate a few reflections on it. The use of advanced techniques causes a clearer detachment of production processes from human work. Robotics and automation were initially supposed to facilitate the production of elements, but over time they began to replace humans more and more. Will there be a place for human work in the future? It is already being pushed out of many bastions in which it was supposed to be indispensable. Autonomous cars and buses are the best example of this. Can similar phenomena be noticed in design? Will the machine replace the creator? The development of artificial intelligence (AI) shows that it is possible. Complicated algorithms are already able to compose a piece of classical music. In the case of architecture, architects are still in the lead, however, one has to take into account the conquest of this field by AI. At the moment, designers have various advanced techniques at their disposal to facilitate and accelerate their work. The most important among them are: digital 3D modeling CAD (Computer Aided Design), Building Information Modeling (BIM), visualizations, and computer animations mainly used to present ready-made ideas, but also useful at the concept stage. Apart from them, three-dimensional printing is also important, as well as three-dimensional design of structures. The above technologies are increasingly used in the design process. They are more compatible with each other than before. They allow you to save labour, accelerate the implementation of tasks, as well as to optimise the designed buildings in many respects related to construction, prefabrication or energy efficiency, to name just a few. An important, although not very common, advantage of technological innovations is their use not only during design and construction, but also during the maintenance of ready-made buildings. The best example of this is BIM, which facilitates the previous management of these technology designed objects. In the future, it will be much easier to design the adaptations of such buildings and to store information of changes which were made. This approach fits in with the idea of Management of Change, which can be included in an even broader aspect among the paradigm of sustainable development.

Technology On Trial: Facilitative And Prejudicial Effects Of Computer-Generated Animations On Jurors' Legal Judgments

The recent emergence of electronic courtrooms (i.e., courtrooms that are equipped with advanced digital technologies) has generated novel ways to present evidence to jurors. Computer-generated animations, which recreate or illustrate the alleged sequence of events in a crime, are increasingly being used by lawyers to present testimonial evidence to jurors. The current study used a 3 (modality: oral vs. static visual vs. animation) x 2 (congruence: incongruent vs. congruent) between-subjects design to investigate whether presentation modality and evidence congruence affect jurors’ ability to properly evaluate evidence and render ‘accurate’ verdicts. In a laboratory setting, mock jurors (N = 238) read a transcript from a fictitious second-degree murder trial. Participants read testimony from eight witnesses, and heard the oral testimony of the defendant with a static visual aid, a computer-generated animation, or no visual aid. Results demonstrated that mock jurors were more likely to acquit the defendant when his testimony was illustrated with a computer-generated animation compared to a static visual aid or with no additional aid. Research in this area can inform the development of evidentiary regulations which adequately govern the admissibility of computer-generated animations in the courtroom, so as ensure that they are used in a way that maintains a defendant’s right to a fair trial. Keywords: computer-generated evidence, computer animations, legal decision-making, information processing, electronic courtrooms

Technology On Trial: Facilitative And Prejudicial Effects Of Computer-Generated Animations On Jurors' Legal Judgments

The recent emergence of electronic courtrooms (i.e., courtrooms that are equipped with advanced digital technologies) has generated novel ways to present evidence to jurors. Computer-generated animations, which recreate or illustrate the alleged sequence of events in a crime, are increasingly being used by lawyers to present testimonial evidence to jurors. The current study used a 3 (modality: oral vs. static visual vs. animation) x 2 (congruence: incongruent vs. congruent) between-subjects design to investigate whether presentation modality and evidence congruence affect jurors’ ability to properly evaluate evidence and render ‘accurate’ verdicts. In a laboratory setting, mock jurors (N = 238) read a transcript from a fictitious second-degree murder trial. Participants read testimony from eight witnesses, and heard the oral testimony of the defendant with a static visual aid, a computer-generated animation, or no visual aid. Results demonstrated that mock jurors were more likely to acquit the defendant when his testimony was illustrated with a computer-generated animation compared to a static visual aid or with no additional aid. Research in this area can inform the development of evidentiary regulations which adequately govern the admissibility of computer-generated animations in the courtroom, so as ensure that they are used in a way that maintains a defendant’s right to a fair trial. Keywords: computer-generated evidence, computer animations, legal decision-making, information processing, electronic courtrooms

Zebrafish Neighbor Distance Changes Relative to Conspecific Size, Position in the Water Column, and the Horizon: A Video-Playback Experiment

Many fish form schools and maintain visual contact with their neighbors in a three-dimensional environment. In this study, we assessed whether zebrafish modified their spacing and interaction time in an additive or multiplicative way relative to multiple sources of social information using computer animations. We simultaneously manipulated: (a) the size of the virtual conspecific (as a proxy of social cue magnitude), (b) the position of the virtual conspecific in the water column (as a proxy of the level of perceived risk), and (c) the absence/presence of the visual horizon (as a proxy of depth perception). We found that the size of the virtual conspecific independently affected spacing behavior (zebrafish increased their separation distance as conspecific size increased). However, some of these factors interacted significantly, such that their effects on social behavior depended on each other. For instance, zebrafish increased their separation distance under high risk conditions when the virtual conspecific was larger, but this risk effect disappeared when the conspecific was the same size or smaller, likely to avoid aggression. Also, zebrafish increased their separation distance when depth perception was enhanced under low risk conditions, but the effect of depth perception disappeared under high risk conditions. Overall, we found that certain dimensions of the visual social environment affected zebrafish spacing behavior in different ways, but they did not affect social interaction time. We discuss the implications of these findings for the spatial organization of fish schools.

EFFECTIVE FORMS OF IN-SERVICE TEACHER TRAINING

Recently, when choosing types of in-service teacher training, the emphasis has shifted towards increasing the activity of participants in the educational process. Therefore, more effective are those types that are based on active interaction and intensive communication, based on unique life and professional experience and willingness to share it. Among such types of in-service teacher training, the authors distinguish participation in trainings, during which attention is paid to the practical development of skills through their own educational experience, and master classes, during which attention is paid to practical demonstration of creative solutions to a problem, new methodology, technology, author's model. The Department of Computer Science of Makarenko Sumy State Pedagogical University developed short courses of in-service teacher (computer science, physics, mathematics) training of Sumy region, which are focused on the formation / development of skills to create computer animations to support the educational process which was organized as trainings and master classes. The pedagogical experiment consisted in a comparative study of the effectiveness of selected learning forms (training or master class) for heterogeneous groups (groups of teachers with different experience). In addition, the following things were determined: degree of use of IT in professional activities, willingness to use computer animation at lessons, self-assessment of skills to develop computer animations for the learning purpose. According to the results of the experiment, it was found that the master class was more effective for teachers with less than 10 years of experience, and training – for teachers with more than 10 years of experience.

Nothobranchius furzeri as an emerging model for mate choice: female choice revealed by animations

The African turquoise killifish (Nothobranchius furzeri) is the shortest-lived vertebrate research model. It is also sexually dimorphic, making it suitable for studying sexual selection. We take advantage of a natural tail colour polymorphism in males and investigate female responses to computer animations of males that differ in this phenotype. Our findings indicate that GRZ (Gonarezhou) females prefer animated males with traits specific to their strain (a yellow tail with a black band) compared to males exhibiting traits from another strain of the same species (a red tail). When females were simultaneously shown animations of both males, they spent significantly more time on the side of the tank where the yellow-tailed animation was visible, and significantly more time interacting with the yellow-tailed animation. Given these repeatable responses and the availability of genomic resources, N. furzeri represents an excellent, untapped model for studying the genetic basis of preferences and reproductive behaviours.

What is the Difference between Online Learning and Traditional Teaching in Medical Education?

The online learning, which is now widely implemented owing to the constraints imposed by COVID-19 pandemic, offers many advantages when compared to the traditional teaching. The online teaching course is available to student any time and can be accessed from any location, providing there is a connection to the Internet, which allows remote learning at home using a fl exible and convenient study schedule that does not disturb other daily activities. The online course contains more informative learning materials, because the multimedia software enables to combine a variety of diverse elements within a given teaching unit, such as texts, PowerPoint presentations, computer animations, video clips, and audio fi les. The amount of time needed to go through the topics of the online course can be variable for different students, depending on their basic knowledge and motivation for learning, thus supporting more personalized learning experience. In contrast to the traditional teaching, that applies a teacher-centered approach, with online teaching the students become independent learners taking the main responsibility for the knowledge acquisition. Therefore, the students’ ability to succeed with online learning is critically determined by the self-discipline, rational time management skills, and monitoring the individual progress. While traditional teaching is largely based on passive learning through memorization of information by student, the online course offers multiple opportunities for self-directed, active learning, including questions for self-assessment, the web links to the relevant learning resources, and the interaction with peers using text messengers, discussion forums and web chats. The weak points of the online learning approach are that the self-studies are easy to procrastinate when the student is not strongly motivated to learn, and that it limits social contacts, leading to the feeling of isolation. Importantly, when developing the online learning course, the computer technology is supposed to be just a technical instrument, whereas the main emphasis is placed on the quality of teaching materials and their relevance to the objectives of the medical curriculum.