Peer Agents Research Articles

Precedent and rest stop convergence in reflective equilibrium

The method of reflective equilibrium is typically characterized as a process of two kinds of adjustments: hold fixed one’s current set of commitments/intuitions and adjust rules/principles to account for them, then hold fixed those rules while making adjustments to one’s set of commitments. Repeat until no further adjustments are required. Such characterizations ignore the role of precedent, i.e., information about the commitments and rules of others and how those might serve as guides in one’s own process of deliberation. In this paper we develop a model that narrows in on a local part of a reflective equilibrium process in which a group of peer agents start with identical lists of commitments and work towards finding the rule that captures that set. If successful, we say they reach rest stop interpersonal convergence. The trouble is that, without precedent, reaching this convergence rests on implausible deterministic assumptions. We analyze how far the inclusion of a kind of soft precedent goes towards aiding the group in reaching said convergence when those assumptions are relaxed.

Peer Incentive Reinforcement Learning for Cooperative Multiagent Games

Social learning, especially social incentives, is extremely important for humans to achieve a high level of coordination. Inspired by this, we introduce this concept into cooperative multi-agent reinforcement learning (MARL), to implicitly address the credit assignment problem and promote the inter-agent direct interactions for cooperations among agents in cooperative multi-agent games. In this paper, we propose a novel Intrinsic Reward method with Peer Incentives (IRPI) based on actor-critic policy gradient. This method can enable agents to incentivize each other for their cooperations through using causal influence among them. Specifically, a novel intrinsic reward mechanism is innovatively designed to empower each agent the ability to give positive or negative rewards to other peer agents' actions through considering the causal influence of the other agents on it. The mechanism is realized by a feed-forward neural network through utilizing causal influence between the agents. The causal influence of one agent on another is inferred via counterfactual reasoning using the joint action-value function in MARL. The quality of the influence is assessed via counterfactual reasoning using the individual value function in MARL. Simulations are carried out on two popular multi-agent game testbeds: Starcraft II Micromanagement and Multi-Agent Particle Environments. Simulational results demonstrate that the proposed IRPI can enhance cooperations among the agents to achieve better performance compared with a number of state-of-the-art MARL methods in a variety of cooperative multi-agent games.

Role design considerations of conversational agents to facilitate discussion and systems thinking

Conversational agents can facilitate learning discussions by applying natural language understanding to process students' discourse. Agents can assume the roles of figures such as peers or mentors, to promote actions similar to human interactions. In this study, we explore how and for whom different role designs of a text-based agent (i.e., chatbot) can facilitate discussion patterns and systems thinking in small-group discussions. Participants included 172 students in 9th grade (ages 13–14). Participants were randomly assigned to groups of five students and interacted with no agent, an expert agent, or a less knowledgeable peer agent. Results suggest that both agents facilitated learning of systems mechanisms by enhancing transactive exchange, where students built on prior ideas in their discussion groups. We also found differences in the agents' effects on discussion and learning outcomes based on groups' variation in systems thinking pre-test. Findings highlight the importance of role design considerations of agents in group settings.

Interacting with educational chatbots: A systematic review

Chatbots hold the promise of revolutionizing education by engaging learners, personalizing learning activities, supporting educators, and developing deep insight into learners’ behavior. However, there is a lack of studies that analyze the recent evidence-based chatbot-learner interaction design techniques applied in education. This study presents a systematic review of 36 papers to understand, compare, and reflect on recent attempts to utilize chatbots in education using seven dimensions: educational field, platform, design principles, the role of chatbots, interaction styles, evidence, and limitations. The results show that the chatbots were mainly designed on a web platform to teach computer science, language, general education, and a few other fields such as engineering and mathematics. Further, more than half of the chatbots were used as teaching agents, while more than a third were peer agents. Most of the chatbots used a predetermined conversational path, and more than a quarter utilized a personalized learning approach that catered to students’ learning needs, while other chatbots used experiential and collaborative learning besides other design principles. Moreover, more than a third of the chatbots were evaluated with experiments, and the results primarily point to improved learning and subjective satisfaction. Challenges and limitations include inadequate or insufficient dataset training and a lack of reliance on usability heuristics. Future studies should explore the effect of chatbot personality and localization on subjective satisfaction and learning effectiveness.

The Square Root Agreement Rule for Incentivizing Truthful Feedback on Online Platforms

A major challenge in obtaining evaluations of products or services on e-commerce platforms is eliciting informative responses in the absence of verifiability. This paper proposes the square root agreement rule (SRA): a simple reward mechanism that incentivizes truthful responses to objective evaluations on such platforms. In this mechanism, an agent gets a reward for an evaluation only if the answer matches that of a peer agent, where this reward is inversely proportional to a popularity index of the answer. This index is defined to be the square root of the empirical frequency at which any two agents performing the same evaluation agree on the particular answer across evaluations of similar entities operating on the platform. Rarely agreed-on answers thus earn a higher reward than answers for which agreements are relatively more common. We show that in the many tasks regime, the truthful equilibrium under SRA is strictly payoff-dominant across large classes of natural equilibria that could arise in these settings, thus increasing the likelihood of its adoption. Although there exist other mechanisms achieving such guarantees, they either impose additional assumptions on the response distribution that are not generally satisfied for objective evaluations or they incentivize truthful behavior only if each agent performs a prohibitively large number of evaluations and commits to using the same strategy for each evaluation. SRA is the first known incentive mechanism satisfying such guarantees without imposing any such requirements. Moreover, our empirical findings demonstrate the robustness of the incentive properties of SRA in the presence of mild subjectivity or observational biases in the responses. These properties make SRA uniquely attractive for administering reward-based incentive schemes (e.g., rebates, discounts, reputation scores, etc.) on online platforms. This paper was accepted by Kartik Hosanagar, information systems. Supplemental Material: The data files and online appendix are available at https://doi.org/10.1287/mnsc.2022.4375 .

Multi-agent-based hybrid peer-to-peer system for distributed information retrieval

With the increasingly huge amount of data located in various databases and the need for users to access them, distributed information retrieval (DIR) has been at the core of the preoccupations of a number of researchers. Indeed, numerous DIR systems and architectures have been proposed including the broker-based architecture. Moreover, providing DIR with more flexibility and adaptability has led researchers thinking to build DIR with software agents. Thus, this research proposes a design and an implementation of a novel system based on the broker-based architecture and the peer-to-peer (P2P) network called broker-based P2P network. The proposed architecture is implemented with a multi-agent system (MAS) where the main agent playing the role of the broker, receives query from a peer agent and forwards them to other peer agents each with their index and resources. Upon completing retrieval process at each peer agent, results are directly sent to the peer agent that initiated the query without using the broker agent. Java Agent DEvelopment framework (JADE) is used to implement the agents and, for experiments, TERRIER (TERabyte RetRIEveR) is extended and used as the search engine to retrieve the Text Retrieval Conference (TREC) collections dataset notably TREC-6. The peer agent that originated the query progressively collects results coming from other peer agents, normalises and merges them and then proceeds with re-ranking. For normalisation, MinMax and Sum that are unsupervised normalisation methods are used.

Teaching and learning with children: Impact of reciprocal peer learning with a social robot on children’s learning and emotive engagement

Pedagogical agents are typically designed to take on a single role: either as a tutor who guides and instructs the student, or as a tutee that learns from the student to reinforce what he/she knows. While both agent-role paradigms have been shown to promote student learning, we hypothesize that there will be heightened benefit with respect to students’ learning and emotional engagement if the agent engages children in a more peer-like way — adaptively switching between tutor/tutee roles. In this work, we present a novel active role-switching (ARS) policy trained using reinforcement learning, in which the agent is rewarded for adapting its tutor or tutee behavior to the child’s knowledge mastery level. To investigate how the three different child–agent interaction paradigms (tutee, tutor, and peer agents) impact children’s learning and affective engagement, we designed a randomized controlled between-subject experiment. Fifty-nine children aged 5–7 years old from a local public school participated in a collaborative word-learning activity with one of the three agent-role paradigms. Our analysis revealed that children’s vocabulary acquisition benefited from the robot tutor’s instruction and knowledge demonstration, whereas children exhibited slightly greater affect on their faces when the robot behaves as a tutee of the child. This synergistic effect between tutor and tutee roles suggests why our adaptive peer-like agent brought the most benefit to children’s vocabulary learning and affective engagement, as compared to an agent that interacts only as a tutor or tutee for the child. This work sheds light on how fixed role (tutor/tutee) and adaptive role (peer) agents support children’s cognitive and emotional needs as they play and learn. It also contributes to an important new dimension of designing educational agents — actively adapting roles based on the student’s engagement and learning needs.

Optimal voltage regulation for distribution networks with multi-microgrids

With the increasing penetration of renewables, microgrid integration has been considered as one of the most promising approaches to enhance the utilizations of various types of energy resources in smart distribution systems. In the near future, with the higher levels of intermittent renewables are envisaged, the distribution network operators will face significant challenges to the control and operation of distribution networks dispersedly. Consequently, the development of effective and motivating ancillary service schemes in a decentralized way for executing real-time control and reducing calculation and communication burden is still in its infancy and needs to be researched. To address this issue, this paper explores an optimal voltage regulation method with the participants of multi-microgrids based on multiple agent systems. Without an arbitration agent, peer agents in the multiple agent system calculate voltage sensitivities by local and neighbourhood measurements only. In this paper, a bi-level game model is proposed for voltage control process. In the upper-level, the distribution network operator searches the reasonable incentive mechanism based on the Stackelberg game. In the lower-level, microgrids make voltage control strategies autonomously based on a static game among microgrids. In the proposed method, microgrids participate in voltage control in distribution networks as ancillary service providers while maximizing their own profits. Meanwhile, the distribution network operator reduces the infrastructure reinforcement and avoids unnecessary renewable energy curtailment. Finally, the feasibility and effectiveness of the proposed method has been demonstrated on a modified IEEE 33-bus system.

Two Heads May be Better than One: Learning from Computer Agents in Conversational Trialogues

BackgroundPedagogical agents are computerized talking heads or embodied animated avatars that help students learn by performing actions and holding conversations with the students in natural language. Dialogues occur between a tutor agent and the student in the case of AutoTutor and other intelligent tutoring systems with natural-language conversation. The agents are adaptive to the students’ actions, verbal contributions, and, in some systems, their emotions (such as boredom, confusion, and frustration).Focus of StudyThis paper explores several designs of trialogues (two agents interacting with a human student) that have been productively implemented for particular students, subject matters, and depths of learning. The two agents take on different roles, but often serve as peers and tutors. There are different trialogue designs that address different pedagogical goals for different classes of students. For example, students can (a) observe vicariously two agents interacting, (b) converse with a tutor agent while a peer agent periodically chimes in, or (c) teach a peer agent while a tutor rescues a problematic interaction. In addition, agents can argue with each other over issues and ask what the human student thinks about the argument.Research DesignTrialogues have been developed for systematic experimental investigations in several studies that measure student impressions, learning gains from pretest to post-test on objective tests, and both cognitive and affective states during learning. The studies compare conditions with different pedagogical principles underlying the trialogues in order to assess the impact of these principles on student impressions, learning, emotions, and other psychological measures. Discourse analyses are performed on the language and actions in the log files in order to assess their impacts on psychological measures.RecommendationsTests of these agent-based systems have shown improvements in learning gains and systematic influences on student emotions. In the future, researchers need to conduct more research to empirically evaluate the psychological impact of different trialogue designs on psychological measures. These trialogue designs range from scripted interactions between agents being observed by the student, to the student helping a fellow peer agent, to the student resolving an argument between two agents. The central question is whether the learning experiences and outcomes show improvement over typical human-computer dialogues (i.e., one human and one tutor agent) and conventional pedagogical interventions.

Chinese male adolescents resisting cigarettes from peers: qualitative research on tactics, perceptions and contextual characteristics

Aims: Cigarette passing and sharing contribute to early smoking onset among Chinese male youth. Refusal efficacy, considered pivotal in smoking prevention, has not been addressed in published research regarding specific refusal tactics and resistant communication behaviours. This focus group study aimed to uncover cigarette resistance responses, cigarette-offering agents’ reactions to refusal, social barriers, and contextual characteristics among targeted Chinese male adolescents. Methods: Twenty focus groups with 7–10 male smoking and nonsmoking students per group from vocational and junior colleges (N = 165) were interviewed in two cities in China. The data were coded and analysed. Findings: Non-confrontational refusal tactics were frequently used. Of the nine identified cigarette refusal categories, consistent and firm declarations of the non-smoking status were considered most effective. Upon rejection, a majority of people offering cigarettes made no further attempts. Most participants expressed neutral attitudes towards cigarette refusals, in contrast with the assumption in the literature that teenagers pervasively dislike such refusals. Conclusions: Health educators likely can use the finding of peer agents’ general neutral (rather than negative) attitudes towards cigarette refusal to relieve teenagers’ apprehension about rejecting cigarettes and thereby help build their refusal confidence. The reported effective strategies should be further tested prior to implementation for training teenagers.

Cigarette Initiation Among Chinese Male Teenagers in Early Smoking Interactions

ABSTRACTTo collect in-depth information regarding cigarette initiation interactions in the early smoking experimentation among male youth in China, twenty focus groups with male teenagers (N = 165) were conducted. Focus group discussions indicated a high prevalence of cigarette initiation among peers, and such initiation often translated into immediate smoking. Of the identified cigarette initiation appeals, “face” and normative pressure appeals were the most difficult to reject. Upon first initiation attempts, more teenagers accepted cigarettes than rejected them. The reasons behind both cigarette initiation and acceptance primarily related to facilitating social interaction. Upon being rejected, initiating teenagers (agents) who insisted on offering cigarettes often gained compliance. Profiles of peer agents were constructed in this study. Key practical implications suggest simultaneously prioritizing efforts to reduce cigarette initiation and encouraging cigarette refusal.

Safe buzz: adaptive security for multi-agents through situational awareness

Multi-agent systems deal with interactive agents that possess some degree of autonomy and cooperate towards the achievement of their goals. Multi-agent systems often need to operate in unsafe environments, and hence there is a need for situational awareness and risk assessment security mechanism that can evaluate agent data to detect threats and analyse the risk posed by them. Basic belief-desire-intention (BDI) architecture lacks a framework that would facilitate effective communication, to bring about situational awareness among agents in an insecure environment. This paper addresses the problems faced by a group of agents lacking global knowledge and global communications, by the introduction of a belief sharing and risk assessment mechanism. We have extended basic BDI architecture with the concept of situational awareness and adaptive risk management. With the proposed architecture, agents can gain awareness by exchanging beliefs, ascertaining the truth of such beliefs and measuring the credibility of a peer agent. Moreover agents can modify their level of alertness by monitoring the risks faced by them and by their peers. This enables the agents to detect and assess risks faced by them in an efficient manner, thereby increasing operational efficiency and resistance against attacks.

Safe buzz: adaptive security for multi-agents through situational awareness

Multi-agent systems deal with interactive agents that possess some degree of autonomy and cooperate towards the achievement of their goals. Multi-agent systems often need to operate in unsafe environments, and hence there is a need for situational awareness and risk assessment security mechanism that can evaluate agent data to detect threats and analyse the risk posed by them. Basic belief-desire-intention (BDI) architecture lacks a framework that would facilitate effective communication, to bring about situational awareness among agents in an insecure environment. This paper addresses the problems faced by a group of agents lacking global knowledge and global communications, by the introduction of a belief sharing and risk assessment mechanism. We have extended basic BDI architecture with the concept of situational awareness and adaptive risk management. With the proposed architecture, agents can gain awareness by exchanging beliefs, ascertaining the truth of such beliefs and measuring the credibility of a peer agent. Moreover agents can modify their level of alertness by monitoring the risks faced by them and by their peers. This enables the agents to detect and assess risks faced by them in an efficient manner, thereby increasing operational efficiency and resistance against attacks.

71 Hypertensive syndrome of risk prevention specific of pregnancy in teens – A technology health

Hypertension in pregnancy is an obstetric injury that usually occurs after the 20th week of gestation or more often characterized by the symptom triad hypertension, edema and proteinuria, in this case, called preeclampsia (PE), not convulsively and eclampsia (EC) convulsively, plus the symptoms of encephalopathy PE, manifesting from the 21th week of pregnancy or within 48 h after delivery, is more common in gilts, thus affecting younger women. Participatory research in order to assess behavioral changes in pregnant adolescents in the prevention and/or control of the GHS risk factors, with the application of educational technology in health (TES). The TES has been prepared based on Belief Model Health and health education presuppositions, whose activities developed in ten educational meetings held by group dynamics, aiming at motivating and strengthening emotional bonds, thereby building collective knowledge about the conduct of prevention and control of the GHS risk factors, and the incorporation of these behaviors in daily life. They participated in the study twenty-five pregnant women assisted in the prenatal care in a Basic Health Unit Family, in Fortaleza, between the first and second trimester of pregnancy, with psycho-emotional conditions to participate in the study. The age of these adolescents ranged between 13 and 19 years. Fifteen were brown, eight white and two black. Twenty-three teenagers living in Fortaleza-Ce and two other municipalities. Two were married eight single and fifteen lived consensually with a partner. The monthly family income ranged from half a minimum wage at the time of the survey (121.22U$). With the implementation of TES, we noticed significant changes in the behavior of pregnant women in relation to their health, including as peer agents of educational activities on the environment in which they lived. However, we realized that there were difficulties in the process of change and some predominated in the motivation for the adoption of health-promoting behaviors among these inherent to the prevention and/or control of the GHS risk factors. The TES allowed to pregnant women, family members, colleagues and other, conditions for reflection on the susceptibility and severity of the GHS, but also happens to mediate the possible changes, encouraging them to decision making and adoption of healthy behaviors. Thus, the TES applied was effective since it helped to make them self-care agents and multipliers of health actions in family and community. Forward to the results achieved, we consider the implementation of TES, based on MCS, as singular validity, thus making it necessary to insert it following the prenatal these women, providing assistance in prenatal, effective and fully paid.

The social dimension of learning through argumentation: Effects of human presence and discourse style.

In spite of its potential for learning, and in particular knowledge revision, argumentation on science concepts is neither easily elicited nor easily sustained. Students may feel uneasy critiquing and being critiqued, especially on complex science topics. We report on a controlled study that tested the role of 2 potential factors that may either relieve or aggravate some of these concerns: the partner’s argumentative discourse style (disputative or deliberative) and belief in interaction with a human or a computer agent. Learners interacted in scripted, computer-mediated interactions with a confederate on their understanding of a scientific concept they had just studied (i.e., diffusion). They were led to believe they were either interacting with a human peer or with a conversational peer agent. The peer confederate’s verbal behavior was scripted to evoke argumentative discourse, while controlling for exposure to conceptual content and the type of dialogue moves, but differed in argumentative discourse style (disputative or deliberative). Results show that conceptual understanding of participants in the deliberative discourse style condition was higher than that in the disputative condition. Further, even though previous studies have reported that the belief in human interaction benefits learning in consensual interactions, the opposite was found to be true in a setting of disagreement and critique: Belief in interaction with a computer agent resulted in higher conceptual learning gains, compared to belief in interaction with a human peer. Implications for theory as well as instructional design are discussed. (PsycINFO Database Record (c) 2016 APA, all rights reserved)

Secure P2P Cross-Community Health Record Exchange in IHE Compatible Systems

This paper investigates a secure mechanism for Electronic Health Records (EHR) exchange over a Peer to Peer (P2P) agent based coordination framework. Our study is based on the SemHealthCoord framework, a platform for the exchange of EHR between autonomous health organisations that extends the existing interoperability standards as proposed by the Integrating Healthcare Enterprise (IHE). Every health organisation in SemHealthCoord represents a community within a P2P network. Communities use a set of autonomous agents and a set of distributed coordination rules to coordinate the agents in the search of specific health records. To enable secure interactions among communities, we propose the use of asymmetric keys and digital certificates. We specify the interaction protocols to provide integrity and authenticity between the communities, and, to illustrate the scalability of our approach, we evaluate the proposed solution in distributed settings by comparing the performance between secured and unsecured data exchange. The contribution of this work is that enables IHE based health communities to dynamically exchange EHR of patients with a secure P2P agent coordination framework.

Innovative Service-Oriented Information System for supporting Mobility in Ubiquitous Spaces: A Role Driven Approach

Conventional architectures based on a central servers are not suitable and cannot scale well in a distributed and high dynamic computing enviroment. Futuristic ubiquitous landscape requires the development of robust, reliable and flexible distributed information systems capable of large scaling and supporting the large amount of simultaneous user requests. In this paper, we propose an innovative information system based on intelligent agents, which is capable of optimizing the search task and managing the services within a coalition of intelligent agents. Users will request a set of information services provided by information providers distributed among a Big Data Network (BDN). The system is divided into two subsystems. The first subsystem is responsible for the search optimization task using an evolutionary algorithm. It generates optimized Workpans for mobile agents that collect needed data from the BDN. The scond subsystem is composed of coalitions of peer agents whitch are endowed with the ability of role switching. A peer agent is by default a passive agent (client), once it receives a given service it can switch its role to become an active agent (service provider) for a given period of time. We provide in the end of this paper simulations supporting our claims about role switching performances and statistical results about the coalition of agents before and after role switching.

The Role of Peers as Agents of Identity Formation in Japanese First-Year University Students

This study demonstrated that peers can act as agents of late adolescent identity formation. A short-term longitudinal study was used to examine identity formation in Japanese first-year university students (n = 170), specifically focusing on identity exploration in the learning sphere among students during their transition to a university. The study examined characteristics of adolescent peer agents and described how they facilitate identity exploration of other adolescents. Five categories of perspectives and actions of peer agents were identified: identity concern, identity goals, identity clues, companionship, and bridging. Peer agents shared learning goals and provided motivation in the process of exploration. Implications of these findings for the role of peers as agents in adolescent identity formation and the characteristics of peer agents in Japanese cultural context are discussed.

Fluctuated peer selection policy and its performance in large-scale multi-agent systems

This paper describes how, in large-scale multi-agent systems, each agent's adaptive selection of peer agents for collaborative tasks affects the overall performance and how this performance varies with the workload of the system and with fluctuations in the agents' peer selection policies (PSP). An intelligent agent in a multi-agent system (MAS) often has to select appropriate agents to assign tasks that cannot be executed locally. These collaborating agents are usually chosen according to their skills. However, if multiple candidate peer agents still remain a more efficient agent is preferable. Of course, its efficiency is affected by the agent' workload and CPU performance and the available communication bandwidth. Unfortunately, as no agent in an open environment such as the Internet can obtain such data from any other agent, this selection must be done according to the available local information about the other known agents. However, this information is limited, usually uncertain and often obsolete. Agents' states may also change over time, so the PSP must be adaptive to some extent. We investigated how the overall performance of MAS would change under adaptive policies in which agents selects peer agents using statistical/reinforcement learning. We particularly focused on mutual interference for selection under different workloads, that is, underloaded, near-critical, and overloaded situations. This paper presents simulation results and shows that the overall performance of MAS highly depends on the workload. It is shown that when agents' workloads are near the limit of theoretical total capability, a greedy PSP degrades the overall performance, even after a sufficient learning time, but that a PSP with a little fluctuation, called fluctuated PSP, can considerably improve it. This paper is the revised and extended version of our conference papers [21] and [22].

Intelligent Peer Networks for Collaborative Web Search

Collaborative query routing is a new paradigm for Web search that treats both established search engines and other publicly available indices as intelligent peer agents in a search network. The approach makes it transparent for anyone to build their own (micro) search engine, by integrating established Web search services, desktop search, and topical crawling techniques. The challenge in this model is that each of these agents must learn about its environment— the existence, knowledge, diversity, reliability, and trustworthiness of other agents — by analyzing the queries received from and results exchanged with these other agents. We present the 6S peer network, which uses machine learning techniques to learn about the changing query environment. We show that simple reinforcement learning algorithms are sufficient to detect and exploit semantic locality in the network, resulting in efficient routing and high-quality search results. A prototype of 6S is available for public use and is intended to assist in the evaluation of different AI techniques employed by the networked agents.