The emergence of cooperation between competing agents has been commonly studied through evolutionary games, but such cooperation often requires a mechanism or a third party to be activated and kept ...alive. To investigate how a mechanism affects the evolution of cooperation, this paper proposes an innovative reinforcement learning-based strategy updating model. The model consists of two symmetrical sets of convolutional neural networks. Besides, the agents’ strategies updating rules are defined: firstly, the agents learn and predict the environment and the behaviors of neighboring agents, then estimate their future payoffs based on this information, and finally determine their strategies based on these estimated payoffs. Through investigating the behavior characteristics and the stable states of the network for highly intelligent agents with memory learning and prediction ability in the evolution of the prisoner’s dilemma game, the results demonstrate that the game initiators who adopt the mixed optimal payoff approach can increase the number of cooperators and facilitate “global cooperation” and “repaying kindness with kindness”. Although the temptation factor has little effect on the population, increasing the discount factor can expand the scale of the cooperative cluster and even achieve dynamic stability. Additionally, a smaller size of minibatch is beneficial for the evolution of cooperation in a smaller experience replay pool. A larger size of minibatch is more conducive to the evolution of cooperation with an increasing capacity of the experience replay pool. This research provides a novel perspective from reinforcement learning to understand the evolution of cooperation.
•An innovative RLSUM is proposed to investigate the PDG.•The behavioral characteristics and the steady state of the network are studied.•The RLSUM can facilitate “global cooperation” and “repaying kindness with kindness”.•The research provides a novel perspective from RL to understand the cooperative evolution.
Protecting people from cyber threats imposes great challenges, not only technically, but also socially. To achieve the intended level of awareness, software security principles need to be shown with ...concrete examples during security education. This study aims to design a serious game integrating software security knowledge and concepts into the processes to make it more engaging to learn while playing. In this paper, we have: (i) designed a serious game to compensate the deficiencies in the literature; (ii) performed empirical evaluations including survey, brainstorming and observation to the proposed game. Results: Our study shows that: (i) Cyber Security-Requirements Awareness Game (CSRAG) has a positive effect on players security learning outcomes, level of engagement and participation; (ii) Game-based learning can be an effective way of teaching security related scenarios.
Internet of Things applications using sensors and actuators raise new privacy related threats, such as drivers and vehicles tracking and profiling. These threats can be addressed by developing ...adaptive and context-aware privacy protection solutions to face the environmental constraints (memory, energy, communication channel, and so on), which cause a number of limitations for applying cryptographic schemes. This paper proposes a privacy preserving solution in ITS context relying on a game theory model between two actors (data holder and data requester) using an incentive motivation against a privacy concession or leading an active attack. We describe the game elements (actors, roles, states, strategies, and transitions) and find an equilibrium point reaching a compromise between privacy concessions and incentive motivation. Finally, we present numerical results to analyze and evaluate the theoretical formulation of the proposed game theory-based model.
A gamification approach for tackling waste management planning and urban development provide a more engaging and interactive experience with high pedagogical potential. Existing serious games ...involving waste management are complex in their data ingestion, use, and presentation, limiting individuals' opportunities to gain knowledge and decision-making skills transferrable to the real world. Simulations, by comparison, provide either an oversimplified and unrealistic user interface or explore in depth individual rather than aggregate key performance indicators for waste management, limiting potential knowledge retention. There is a clear opportunity in creating an informative, easy-to-use simulation-based game to help stakeholders build understanding of waste management policies, performance, and causal relationships. This gamified tool provides clear feedback through quick-visibility performance indicators (i.e., waste accumulation index, waste compositional analysis, prevention activities etc.) and offers the opportunity, through multi-criteria decision making, of simulating real-life scenarios and previewing the possible outcomes of certain in-game actions. The research question is how the process of gamification might serve as powerful tool for educating decision makers. The results are considered as a reference point to any policy maker intending to assess environmental performance, proposed activities to reach Circular Economy targets, and European Green Deal and UN Sustainable Development Goals.
Display omitted
•Gamification tools for waste management strategic development•Overview of game-based learning applications•Gamification for pursuing Circular Economy, UNSDGs, EGD•Integration of KPIs to gamification for urban waste management planning
CONVERGENCE TO THE MEAN FIELD GAME LIMIT Nutz, Marcel; San Martin, Jaime; Tan, Xiaowei
The Annals of applied probability,
02/2020, Volume:
30, Issue:
1
Journal Article
Peer reviewed
Open access
We study the convergence of Nash equilibria in a game of optimal stopping. If the associated mean field game has a unique equilibrium, any sequence of n-player equilibria converges to it as n → ∞. ...However, both the finite and infinite player versions of the game often admit multiple equilibria. We show that mean field equilibria satisfying a transversality condition are limit points of n-player equilibria, but we also exhibit a remarkable class of mean field equilibria that are not limits, thus questioning their interpretation as “large n” equilibria.
Evidence indicates that downloading on-demand videos accounts for a dramatic increase in data traffic over cellular networks. Caching popular videos in the storage of small-cell base stations (SBS), ...namely, small-cell caching, is an efficient technology for reducing the transmission latency while mitigating the redundant transmissions of popular videos over back-haul channels. In this paper, we consider a commercialized small-cell caching system consisting of a network service provider (NSP), several video retailers (VRs), and mobile users (MUs). The NSP leases its SBSs to the VRs for the purpose of making profits, and the VRs, after storing popular videos in the rented SBSs, can provide faster local video transmissions to the MUs, thereby gaining more profits. We conceive this system within the framework of Stackelberg game by treating the SBSs as specific types of resources. We first model the MUs and SBSs as two independent Poisson point processes, and develop, via stochastic geometry theory, the probability of the specific event that an MU obtains the video of its choice directly from the memory of an SBS. Then, based on the probability derived, we formulate a Stackelberg game to jointly maximize the average profit of both the NSP and the VRs. In addition, we investigate the Stackelberg equilibrium by solving a non-convex optimization problem. With the aid of this game theoretic framework, we shed light on the relationship between four important factors: the optimal pricing of leasing an SBS, the SBSs allocation among the VRs, the storage size of the SBSs, and the popularity distribution of the VRs. Monte Carlo simulations show that our stochastic geometry-based analytical results closely match the empirical ones. Numerical results are also provided for quantifying the proposed game-theoretic framework by showing its efficiency on pricing and resource allocation.
•We study cultural differences in strength of social norms for cooperation and coordination.•We study the effect of societal threats on the evolved strength of such social norms in evolutionary ...models.•We find higher threat leads to stronger norms and more punishment of deviance.•Our results illuminate the evolutionary basis for cultural variation in norm strength.
The strengths of social norms vary considerably across cultures, yet little research has shown whether such differences have an evolutionary basis. Integrating research in cross-cultural psychology with evolutionary game theory, we show that groups that face a high degree of threat develop stronger norms for organizing social interaction, with a higher degree of norm–adherence and higher punishment for deviant behavior. Conversely, groups that have little threat can afford to have weaker norms with less punishment for deviance. Our results apply to two kinds of norms: norms of cooperation, in which individuals must choose whether to cooperate (thereby benefitting everyone) or enrich themselves at the expense of others; and norms of coordination, in which there are several equally good ways for individuals to coordinate their actions, but individuals need to agree on which way to coordinate. This is the first work to show that different degrees of norm strength are evolutionarily adaptive to societal threat. Evolutionary game theoretic models of cultural adaptation may prove fruitful for exploring the causes of many other cultural differences that may be adaptive to particular ecological and historical contexts.
Computational thinking (CT) education has drawn increasing attention from educators and researchers. This study conducted a meta-analysis of 27 empirical studies to examine the effectiveness of ...game-based learning (GBL) for fostering students’ CT. The effects of various factors on the learning process for acquiring CT were also examined. The results showed that (a) conducting GBL can foster students’ CT, and the overall effect was at the upper-middle level (Hedges’ g = .600, 95% CI .465, .735, p < .001). (b) Furthermore, conducting GBL can improve students’ CT concepts (Hedges’ g = .916, 95% CI .410, 1.423, p < .001), CT skills (Hedges’ g = .494, 95% CI .389, .600, p < .001), and CT perspectives (Hedges’ g = .927, 95% CI .039, 1.816, p < .05). (c) Additionally, game mode, teaching context, and participant characteristics have positive effects on CT. Based on the findings, it is suggested that using more unplugged games and video games, designing collaborative game activities, and tailoring approaches according to gender difference and programming experience can effectively promote CT. The results have significance for fostering students’ CT in GBL; it is further suggested that instruction processes be rationally designed.
Previous research shows that digital game‐based learning (DGBL) can have positive effects on engagement, motivation and learning, and that using narratives may reinforce these effects. A systematic ...review identified 15 DGBL systems that report effects from their use of narratives. A gap in the field, however, is the lack of a common model to categorize and isolate narratives in DGBL to enable an analysis and comparison of how, and under what conditions, narratives have effects on learning in DGBL systems. The ludo narrative variable model (LNVM) that has been used to isolate and categorize narratives in research on commercial video games is a candidate to fill this gap. This research has investigated the potential of this model for DGBL and resulted in an extended LNVM (eLNVM) that can be used to isolate and categorize narratives in DGBL. The 15 DGBL systems were categorized on the eLNVM and the results show that there are characteristics of DGBL systems with positive self‐reported effects that separate them from other DGBL systems. Furthermore, it was possible to identify characteristics of the narrative modeling that are associated with positive effects on engagement, motivation and learning. The paper concludes with a description of how the eLNVM will be used in future research.
Polymatrix games belong to a class of multi-player games, in which players interact pairwisely and the underlying pairwise interactions are defined by a simple undirected graph where all the edges ...are completely deterministic. But the link uncertainty between players is not taken into consideration in a standard polymatrix game. In this paper, we put our attention to a special class of polymatrix games — zero-sum polymatrix games, and aim to investigate zero-sum polymatrix games with uncertain links. By considering the diversity of uncertainty, we utilize Dempster-Shafer evidence theory to express the link uncertainty in the games. Then, based on a generalized minmax theorem, we develop a new linear programming model with two groups of constraints to calculate the equilibrium payoffs of players and find the equilibria of the zero-sum plymatrix games with belief links. In terms of these, we also establish a Dempster-Shafer theory solution to zero-sum polymatrix games with link uncertainty. Finally, a numerical example is given to illustrate the potential applications of the proposed model.