MAGNet: Multi-agent Graph Network for Deep Multi-agent Reinforcement Learning
We present two examples of how human-like behavior can be implemented in a model of computer player to improve its characteristics and decision-making patterns in video game. At first, we describe a reinforcement learning model, which helps to choose the best weapon depending on reward values obtained from shooting combat situations. Secondly, we consider an obstacle avoiding path planning adapted to the tactical visibility measure. We describe an implementation of a smoothing path model, which allows the use of penalties (negative rewards) for walking through ``bad'' tactical positions. We also study algorithms of path finding such as improved I-ARA* search algorithm for dynamic graph by copying human discrete decision-making model of reconsidering goals similar to Page-Rank algorithm. All the approaches demonstrate how human behavior can be modeled in applications with significant perception of intellectual agent actions.
In this article a combination of two modern aspects of games development is considered: (i) the impact of high quality graphics and virtual reality (VR) user adaptation to believe in realness of in-game events by user’s own eyes; (ii) modeling an enemy’s behavior under automatic computer control, called BOT, which reacts similarly to human players. We consider a First-Person Shooter (FPS) game genre, which simulates an experience of combat actions. We describe some tricks to overcome simulator sicknesses in a shooter with respect to Oculus Rift and HTC Vive headsets. We created a BOT model that strongly reduces the conflict and uncertainty in matching human expectations. BOT passes VR game Alan Turing test with 80% threshold of believable human-like behavior.
Efficient regulation of internal homeostasis and defending it against perturbations requires adaptive behavioral strategies. However, the computational principles mediating the interaction between homeostatic and associative learning processes remain undefined. Here we use a definition of primary rewards, as outcomes fulfilling physiological needs, to build a normative theory showing how learning motivated behaviors may be modulated by internal states. Within this framework, we mathematically prove that seeking rewards is equivalent to the fundamental objective of physiological stability, defining the notion of physiological rationality of behavior. We further suggest a formal basis for temporal discounting of rewards by showing that discounting motivates animals to follow the shortest path in the space of physiological variables toward the desired setpoint. We also explain how animals learn to act predictively to preclude prospective homeostatic challenges, and several other behavioral patterns. Finally, we suggest a computational role for interaction between hypothalamus and the brain reward system.
The Autonomous Agents and MultiAgent Systems (AAMAS) conference series brings together researchers from around the world to share the latest advances in the field. It is the premier forum for research in the theory and practice of autonomous agents and multi-agent systems. AAMAS 2002, the first of the series, was held in Bologna, followed by Melbourne (2003), New York (2004), Utrecht (2005), Hakodate (2006), Honolulu (2007), Estoril (2008), Budapest (2009), Toronto (2010), Taipei (2011), Valencia (2012), Saint Paul (2013), Paris (2014), and Istanbul (2015). This volume constitutes the proceedings of AAMAS 2016, the fifteenth conference in the series, held in Singapore in May 2016.
In line with previous editions, AAMAS 2016 attracted submissions for a general track and five special tracks: Innovative Applications, Robotics, Embodied Virtual Agents and Human-Agent Interaction, Blue Sky Ideas track, and the JAAMAS presentation track. The special tracks were chaired by leading researchers in their corresponding fields: Onn Shehory and Noa Agmon chaired the Innovative Applications track, Francesco Amigoni and Roderich Gross the Robotics track, Tim Bickmore and Hannes Vilhjálmsson the Embodied Virtual Agents and Human-Agent Interaction track, and Frank Dignum the Blue Sky Ideas track. As a new initiative, the chairs of AAMAS 2016 also solicited articles published in the Journal of Autonomous Agents and Multiagent Systems for the JAAMAS Presentation Track. Only papers that have appeared in the Journal of Autonomous Agents and Multi-agent Systems (JAAMAS) in the 12 months period preceding the AAMAS notification date were eligible. This new track was chaired by Peter Stone.
Jointly with the PC chairs the special track chairs were responsible for appointing the Programme Committee (PC) members and the Senior Programme Committee members (SPC) for their tracks, and they made acceptance/rejection recommendations for their tracks in consultation with Programme Chairs based on input provided by the track PC, SPC, and Area Experts. This year the PC chairs introduced the new role of Area Experts, i.e., SPC members with additional responsibilities, to assist with selecting SPC members for specific research areas, identifying appropriate keywords, and assisting in potential issues during discussion phase. This new role was a success and increased the quality of our SPC and PC, and also the reviewing process in general.
Full paper submissions (8 pages plus bibliographic references) and Blue Sky Ideas paper submissions (4 pages plus references) were solicited for AAMAS 2016. Some of the full paper submissions were accepted as extended abstracts (2 pages). The papers were selected by means of a thorough review and discussion process, which included an opportunity for authors to respond to reviewer comments during a rebuttal phase. All SPC members, Area Experts, and Track Chairs followed and contributed to the technical discussions on the papers they were overseeing. The JAAMAS presentation Track submissions published as extended abstracts were handled by the track chair.
Overall, out of 550 submissions, 137 (25%) were accepted as full papers and 143 (26%) were accepted as extended abstracts. Additionally, all 16 JAAMAS track submissions were accepted.
Full papers were presented orally in 20 minute slots; all extended abstracts and, optionally, full papers were presented as posters during the conference.
Out of the 550 submissions, 351 (64%) had a student as the primary author, 82 of these were accepted as full papers (23%), and a further 90 (26%) were accepted as extended abstracts.
The proceedings also contain 17 Demonstration papers, 13 Doctoral Consortium papers, as well as abstracts of the invited talks and details of some of the awards given.
Constant changes in demand for resources in the market complicate planning and management of material flows. In current practice, it’s possible to solve this problem by applying multi-agent systems representing a set of interacting software objects called intelligent agents. The activity of an intelligent agent is directed at achieving individual goals, which may include the search of possibilities of delivery, storing goods, transportation of goods and other. The article considers models, technologies, the typical architecture of multi-agent system, analyzes the completed projects and describes the prospects for the development of multi-agent systems in logistics.
Humans often change their beliefs or behavior due to the behavior or opinions of others. This study explored, with the use of human event-related potentials (ERPs), whether social conformity is based on a general performance-monitoring mechanism. We tested the hypothesis that conflicts with a normative group opinion evoke a feedback-related negativity (FRN) often associated with performance monitoring and subsequent adjustment of behavior. The experimental results show that individual judgments of facial attractiveness were adjusted in line with a normative group opinion. A mismatch between individual and group opinions triggered a frontocentral negative deflection with the maximum at 200 ms, similar to FRN. Overall, a conflict with a normative group opinion triggered a cascade of neuronal responses: from an earlier FRN response reflecting a conflict with the normative opinion to a later ERP component (peaking at 380 ms) reflecting a conforming behavioral adjustment. These results add to the growing literature on neuronal mechanisms of social influence by disentangling the conflict-monitoring signal in response to the perceived violation of social norms and the neural signal of a conforming behavioral adjustment.
The Smart Innovation, Systems and Technologies book series encompasses the topics of knowledge, intelligence, innovation and sustainability. The aim of the series is to make available a platform for the publication of books on all aspects of single and multi-disciplinary research on these themes in order to make the latest results available in a readily-accessible form. Volumes on interdisciplinary research combining two or more of these areas is particularly sought.
The series covers systems and paradigms that employ knowledge and intelligence in a broad sense. Its scope is systems having embedded knowledge and intelligence, which may be applied to the solution of world problems in industry, the environment and the community. It also focusses on the knowledge-transfer methodologies and innovation strategies employed to make this happen effectively. The combination of intelligent systems tools and a broad range of applications introduces a need for a synergy of disciplines from science, technology, business andthe humanities. The series will include conference proceedings, edited collections, monographs, handbooks, reference books, and other relevant types of book in areas of science and technology where smart systems and technologies can offer innovative solutions.
High quality content is an essential feature for all book proposals accepted for the series. It is expected that editors of all accepted volumes will ensure that contributions are subjected to an appropriate level of reviewing process and adhere to KES quality principles.
Our decisions are affected not only by objective information about the available options but also by other people. Recent brain imaging studies have adopted the cognitive neuroscience approach for studying the neural mechanisms of social influence. A number of studies have shown that social influence is associated with neural activity in the medial prefrontal cortex and ventral striatum, which are two brain areas involved in the fundamental and not exclusively social mechanisms of performance monitoring. Therefore, the neural mechanisms of social influence could be deeply integrated into our general neuronal performance-monitoring mechanisms.