Agent Framework Research Articles

The impact of inclusive fashion advertising with plus-size models on female consumers: The mediating role of brand warmth

ABSTRACT This study investigates the effectiveness of inclusive advertising featuring both plus-size models and straight-size models on consumers’ responses. By building on the Brands as Intentional Agents Framework (BIAF), the study also investigates the mechanism of the positive effect of including plus-size models on brand performance. For this study, a between-subjects design of model body size (diverse versus straight only) and a moderator of consumer body size (plus versus straight) were used; a multivariate analysis of covariance was conducted, along with a mediation analysis, to test the proposed hypotheses. The results indicated that inclusive advertisements depicting models with diverse body sizes had a positive main effect on brand attitude and an indirect effect on brand attitude and purchase intention. Brand warmth mediated the positive effect of diverse body sizes on brand attitude and purchase intention. This study contributes to the literature on body image by demonstrating the effectiveness of including plus-size models on two levels of consumer responses – brand attitude and purchase intention – and by demonstrating the mechanism of brand warmth. This study also informs brand managers as to how embracing plus-size models can benefit brand performance.

Toward a Competitive Agent Framework for Magic: The Gathering

Magic: The Gathering is a collectible card game that is traditionally a human-based game: two or more players compete with their own custom deck to eliminate their opponent(s). Magic has again surged in popularity with the release of Magic: Arena, a digital substitute for the physical card game. What Magic: Arena lacks is meaningful gameplay with a competitive, intelligent agent. We propose a strategy-based framework for basic decision making in Magic: The Gathering. We describe these procedures and provide experimental evidence for the effectiveness of our approach.

DDPG Agent to Swing Up and Balance Cart- Pole System

Reinforcement learning is a mathematical framework for agents to interact intelligently with their environment. Unlike supervised learning, where a system learns with the help of labeled data, reinforcement learning agents learn how to act by trial and error only receiving a reward signal from their environments. A field where reinforcement learning has been prominently successful is robotics [3]. However, real-world control problems are also particularly challenging because of the noise and high- dimensionality of input data (e.g., visual input). In recent years, in the field of supervised learning, deep neural networks have been successfully used to extract meaning from this kind of data. Building on these advances, deep reinforcement learning was used to solve complex problems like Atari games and Go. Mnih et al. [1] built a system with fixed hyper parameters able to learn to play 49 different Atari games only from raw pixel inputs. However, in order to apply the same methods to real-world control problems, deep reinforcement learning has to be able to deal with continuous action spaces. Discretizing continuous action spaces would scale poorly, since the number of discrete actions grows exponentially with the dimensionality of the action. Furthermore, having a parametrized policy can be advantageous because it can generalize in the action space. Therefore with this thesis we study state-of-the-art deep reinforcement learning algorithm, Deep Deterministic Policy Gradients. We provide a theoretical comparison to other popular methods, an evaluation of its performance, identify its limitations and investigate future directions of research. The remainder of the thesis is organized as follows. We start by introducing the field of interest, machine learning, focusing our attention of deep learning and reinforcement learning. We continue by describing in details the two main algorithms, core of this study, namely Deep Q-Network (DQN) and Deep Deterministic Policy Gradients (DDPG). We then provide implementatory details of DDPG and our test environment, followed by a description of benchmark test cases. Finally, we discuss the results of our evaluation, identifying limitations of the current approach and proposing future avenues of research.

Intelligent agent based framework to augment warehouse management systems for dynamic demand environments

Warehouses are being impacted by increasing e-commerce and omni-channel commerce. The design of current WMSs (Warehouse Management Systems) may not be suitable to this mode of operation. The golden rule of material handling is smooth product flow, but there are day-to-day operational issues that occur in the warehouse that can impact this and order fulfilment, resulting in disruptions. Standard operational process is paramount to warehouse operational control but may preclude a dynamic response to real-time operational constraints. The growth of IoT (Internet of Things) sensor and data analytics technology provide new opportunities for designing warehouse management systems that detect and reorganise around real-time constraints to mitigate the impact of day-to-day warehouse operational issues. This paper presents the design and development stage of a design science methodology of an intelligent agent framework for basic warehouse management systems. This framework is distributed, is structured around operational constraints and includes the human operator at operational and decision support levels. An agent based simulation was built to demonstrate the viability of the framework.

Testing an open app-based water management information system and its uptake among stakeholders in the Upper-Comoé river basin (Burkina Faso)

ABSTRACT Managing sustainably natural resources in an increasingly populated world is challenging, even more so when information asymmetries and conflicting interests exacerbate tensions among stakeholders. We used a principal–agent framework to analyse these dilemmas while testing the development of an open-source information platform for water users in the Upper-Comoé river basin (Burkina Faso). Information on water retrieval and delivery was shared through schematic maps in a WhatsApp water user group. By reducing information asymmetries, the system enhanced water delivery and distribution, and attenuated water-related conflicts. The study suggests that an open-source, low-cost application can enhance information access and sharing, reducing common principal–agent problems.

Beyond Objects in Space-Time: Towards a Movement Analysis Framework with ‘How’ and ‘Why’ Elements

Current spatiotemporal data has facilitated movement studies to shift objectives from descriptive models to explanations of the underlying causes of movement. From both a practical and theoretical standpoint, progress in developing approaches for these explanations should be founded on a conceptual model. This paper presents such a model in which three conceptual levels of abstraction are proposed to frame an agent-based representation of movement decision-making processes: ‘attribute,’ ‘actor,’ and ‘autonomous agent’. These in combination with three temporal, spatial, and spatiotemporal general forms of observations distinguish nine (3 × 3) representation typologies of movement data within the agent framework. Thirdly, there are three levels of cognitive reasoning: ‘association,’ ‘intervention,’ and ‘counterfactual’. This makes for 27 possible types of operation embedded in a conceptual cube with the level of abstraction, type of observation, and degree of cognitive reasoning forming the three axes. The conceptual model is an arena where movement queries and the statement of relevant objectives takes place. An example implementation of a tightly constrained spatiotemporal scenario to ground the agent-structure was summarised. The platform has been well-defined so as to accommodate different tools and techniques to drive causal inference in computational movement analysis as an immediate future step.

Recommendation agents and information sharing through social media for coronavirus outbreak

The novel outbreak of coronavirus disease (COVID-19) was an unexpected event for tourism in the world as well as tourism in the Netherlands. In this situation, the travelers’ decision-making for tourism destinations was heavily affected by this global event. Social media usage has played an essential role in travelers’ decision-making and increased the awareness of travel-related risks from the COVID-19 outbreak. Online consumer media for the outbreak of COVID-19 has been a crucial source of information for travelers. In the current situation, tourists are using electronic word of mouth (eWOM) more and more for travel planning. Opinions provided by peer travelers for the outbreak of COVID-19 tend to reduce the possibility of poor decisions. Nevertheless, the increasing number of reviews per experience makes reading all feedback hard to make an informed decision. Accordingly, recommendation agents developed by machine learning techniques can be effective in the analysis of such social big data for the identification of useful patterns from the data, knowledge discovery, and real-time service recommendations. The current research aims to adopt a framework for the recommendation agents through topic modeling to uncover the most important dimensions of COVID-19 reviews in the Netherland forums in TripAdvisor. This study demonstrates how social networking websites and online reviews can be effective in unexpected events for travelers’ decision making. We conclude with the implications of our study for future research and practice.

Identifying key aspects of household behavior in a representative agent framework

We study three key aspects of household behavior, i.e. the form of household preferences, the existence of interdependence between consumption and leisure choices, and the Frisch elasticity of labor supply. We find that an appropriate form of household preferences to use in non-recursive utility functions that incorporates the feature of interdependence is of the Cobb-Douglas type. By solving the household maximization problem, we derive a simultaneous system of two equations, a consumption equation and a labor supply equation, and estimate them with non-linear GMM. Regarding our estimates, the weight of consumption in the utility function, which is on the upper side of the relevant estimates in the literature, and the elasticity of intertemporal substitution seem to drive up the Frisch elasticity value. Our work highlights the importance of both the choice of the form of household preferences and the appropriate estimation procedure in the household maximization problem for obtaining reasonable values of the preference parameters, especially the Frisch elasticity. Our results may be useful for researchers working with the calibration of preference parameters in DSGE models.

Repositioning luxury fashion brands as intentional agents: the roles of consumer envy and admiration

PurposeBuilding on the Brands as Intentional Agents Framework (BIAF), the aim of this study is to demonstrate the effectiveness of social media marketing (SMM) as a tool to communicate luxury fashion brands' good intentions toward the general public.Design/methodology/approachA sample of 488 US female consumers was used to test a conceptual model delineating the sequential linkages from luxury fashion brands' intentions to brand emotions (i.e. envy vs admiration) and to consumer–brand relationships (i.e. emotional brand attachment and brand forgiveness). Structural equation modeling (SEM) was performed to test the measurement and structural models.FindingsThe results indicated that luxury fashion brands' “populist” intentions had a positive impact on consumer admiration. Both consumer envy and admiration had positive effects on emotional brand attachment and brand forgiveness. However, admiration had a stronger effect than envy on these relational consumer responses.Originality/valueThis study identified that luxury fashion brands, frequently stereotyped as exclusive, can become brands admired by mass-market consumers by expressing warmth on social media. Drawing on social psychological perspectives and the BIAF, this study adds to the literature on luxury brands' social media communication by demonstrating the effectiveness of brand warmth to induce consumers' strong relational outcomes.

A trade-credit-based incentive mechanism for a risk-averse retailer with private information

In a two-tier supply chain composed of a risk-neutral supplier and a risk-averse retailer, we consider the retailer’s sales cost to be private information and study the incentive effect of trade credit under information asymmetry. The model is formulated based on the principal–agent framework. We obtain the optimal trade credit contract configuration and further derive the optimal decision of the retailer, which are then compared with the results obtained from the benchmark case with symmetric information. We also analyze the influence of the retailer’s private information and risk-aversion behavior on the contract parameters and the retail price. The study shows that when the retailer’s risk aversion coefficient is within a certain range, the reasonable trade credit contract designed by the supplier can effectively encourage the retailer to report her real sales cost, and increase the supply chain profitability. Finally, numerical examples validate the theoretical model, and illustrate the impacts of private information and risk aversion on the profits of supply chain members and the system.

A Conversational Agent Framework with Multi-modal Personality Expression

Consistently exhibited personalities are crucial elements of realistic, engaging, and behavior-rich conversational virtual agents. Both nonverbal and verbal cues help convey these agents’ unseen psychological states, contributing to our effective communication with them. We introduce a comprehensive framework to design conversational agents that express personality through non-verbal behaviors like body movement and facial expressions, as well as verbal behaviors like dialogue selection and voice transformation. We use the OCEAN personality model, which defines personality as a combination of five orthogonal factors of openness, conscientiousness, extraversion, agreeableness, and neuroticism. The framework combines existing personality expression methods with novel ones such as new algorithms to convey Laban Shape and Effort qualities. We perform Amazon Mechanical Turk studies to analyze how different communication modalities influence our perception of virtual agent personalities and compare their individual and combined effects on each personality dimension. The results indicate that our personality-based modifications are perceived as natural, and each additional modality improves perception accuracy, with the best performance achieved when all the modalities are present. We also report some correlations for the perception of conscientiousness with neuroticism and openness with extraversion.

Bottom-up multi-agent reinforcement learning by reward shaping for cooperative-competitive tasks

A multi-agent system (MAS) is expected to be applied to various real-world problems where a single agent cannot accomplish given tasks. Due to the inherent complexity in the real-world MAS, however, manual design of group behaviors of agents is intractable. Multi-agent reinforcement learning (MARL), which is a framework for multiple agents in the same environment to learn their policies adaptively by using reinforcement learning, would be a promising methodology for such complexity in the MAS. To acquire the group behaviors by MARL, all the agents are required to understand how to achieve the respective tasks cooperatively. So far, we have proposed “bottom-up MARL”, which is a decentralized system to manage real and large-scale MARL, with a reward shaping algorithm to represent the group behaviors. The reward shaping algorithm, however, assumes that all the agents are in cooperative relationships to some extent. In this paper, therefore, we extend this algorithm to allow the agents not to know the interests between them. The interests are regarded as correlation coefficients derived from the agents’ rewards, which are numerically estimated in an online manner. Actually, in both simulations and real experiments without knowledge of the interests between the agents, they correctly estimated their interests, thereby allowing them to derive their new rewards to represent the feasible group behaviors in the decentralized manner. As a result, our extended algorithm succeeded in acquiring the group behaviors from cooperative tasks to competitive tasks.

Robust Deep Reinforcement Learning for Traffic Signal Control

A traffic signal is a fundamental part of the traffic control system to reduce congestion and enhance safety. Since the inception of motorized vehicles, traffic signal controllers are put in place to coordinate and maintain traffic flow. With the number of vehicles on the road increasing exponentially, it is imperative to innovate new traffic control frameworks to cope with the high-density traffic demand. In this regard, recent advances in machine/deep learning have enabled significant progress towards reducing congestion using reinforcement learning for traffic signal control. However, most of these works are still not ready for deployment due to assumptions of perfect knowledge of the traffic environment. In reality, congestion detection or prediction systems are at best able to approximate the traffic state with significant noise. In this work, we propose a robust training framework for reinforcement learning agents that can handle such noisy approximation of the traffic states. Specifically, we show that by carefully adding synthetic perturbations to the state space, such as the queue length during training, the reinforcement learning agents can be robustified. Conceptually, our approach is similar to adversarial training schemes and can lead to successful deployment of reinforcement learning agent-based traffic signal controllers.

Construction of a cloud-based mold manufacturing system based on JAFMAS

Molds are the basis of manufacturing of other products, and thus have higher requirements and more complex design processes than other products. Therefore, studying the technologies used in mold production is necessary. This paper aims to build a cloud manufacturing system of mold manufacturing enterprises based on the Java-based agent framework for multi-agent systems (JAFMAS). Specifically, according to the operation characteristics of different cloud-based manufacturing enterprises, this paper first made in-depth analysis of the cloud manufacturing and multi-agent technology, and put forward a hierarchical model of cloud manufacturing system based on multiple agents. Then, as per the structure of the agents and JAFMAS, the interaction and collaboration mechanism between agents in the system were analyzed to establish the functions of the cloud-based mold manufacturing system, thereby building a cloud-based manufacturing system for mold manufacturers based on JAFMAS.

Agent-based bidirectional charging algorithms for battery electric vehicles in renewable energy communities

This paper introduces a framework for agent based autonomous charging and discharging of Battery Electric Vehicle (BEV) at local energy communities. Agents are programmed to control the bidirectional charging according to green energy utilisation incentives, based on load and generation forecasts. The optimization is achieved within a group of independent prosumers following a fully distributed approach using multiple self-organising agents. No central instance is needed for communication, billing or decision making. To demonstrate the ecological benefits of the system, simulations for a car pool, with a Photovoltaic (PV) plant, in a residential neighborhood were performed. The simulations resulted in an increased community PV self-consumption value of 48% compared to 29% in case of uncontrolled charging processes.

The paradox of regulatory discretion

AbstractRegulatory authorities in the utilities sector typically employ economic evidence and analysis to make expert discretionary judgments under uncertainty. However, economic analysis does not provide clear answers regarding policy outcomes. This exposes regulators to environmental uncertainty, that is, uncertainty regarding the reactions of other actors in the institutional system to their decisions. When policy outcome and environmental uncertainty are high, discretion takes center stage. Will regulators pursue the course of action suggested by economic analysis and their expert judgment or not? What explains this choice? To answer these questions, we carry out a comparative analysis of three British regulatory authorities in the utilities sector: the Office of Communications, the Office of Gas and Electricity, and the Water Services Regulation Authority. We consider two key sectoral and organizational characteristics: the extent of market competition, and statutory discretion. We rely on interview evidence and documentary analysis and a principal–agent framework. Our analysis reveals a paradox: when environmental and policy outcome uncertainty are high, the higher the regulatory discretion, the lower the role of economic expertise in regulatory decisions. Our findings call for a normative reflection on the role of expertise in regulated sectors.

A MULTI-AGENT APPROACH TO POMDPS USING OFF-POLICY REINFORCEMENT LEARNING AND GENETIC ALGORITHMS

This paper introduces novel concepts for accelerating learning in an off-policy reinforcement learning algorithm for Partially Observable Markov Decision Processes (POMDP) by leveraging multiple agents frame work. Reinforcement learning (RL) algorithm is considerably a slow but elegant approach to learning in an unknown environment. Although the action-value (Q-learning) is faster than the state-value, the rate of convergence to an optimal policy or maximum cumulative reward remains a constraint. Consequently, in an attempt to optimize the learning phase of an RL problem within POMD environment, we present two multi-agent learning paradigms: the multi-agent off-policy reinforcement learning and an ingenious GA (genetic Algorithm) approach for multi-agent offline learning using feedforward neural networks. At the end of the trainings (episodes and epochs) for reinforcement learning and genetic algorithm respectively, we compare the convergence rate for both algorithms with respect to creating the underlying MDPs for POMDP problems. Finally, we demonstrate the impact of layered resampling of Monte Carlo’s particle filter for improving the belief state estimation accuracy with respect to ground truth within POMDP domains. Initial empirical results suggest practicable solutions.

A Review Paper on “Modeling of Traffic Simulation System for Dynamic Traffic with Multi Agent Framework”

A Review Paper on "Modeling of Traffic Simulation System for Dynamic Traffic with Multi Agent Framework" - written by Sushant Madhukar Gajbhiye , Dr. Arun Patel published on 2020/09/16 download full article with reference data and citations

ResIoT: An IoT social framework resilient to malicious activities

The purpose of the next internet of things (IoT) is that of making available myriad of services to people by high sensing intelligent devices capable of reasoning and real time acting. The convergence of IoT and multi-agent systems (MAS) provides the opportunity to benefit from the social attitude of agents in order to perform machine-to-machine (M2M) cooperation among smart entities. However, the selection of reliable partners for cooperation represents a hard task in a mobile and federated context, especially because the trustworthiness of devices is largely unreferenced. The issues discussed above can be synthesized by recalling the well known concept of social resilience in IoT systems, i.e., the capability of an IoT network to resist to possible attacks by malicious agent that potentially could infect large areas of the network, spamming unreliable information and/or assuming unfair behaviors. In this sense, social resilience is devoted to face malicious activities of software agents in their social interactions, and do not deal with the correct working of the sensors and other information devices. In this setting, the use of a reputation model can be a practicable and effective solution to form local communities of agents on the basis of their social capabilities. In this paper, we propose a framework for agents operating in an IoT environment, called ResIoT, where the formation of communities for collaborative purposes is performed on the basis of agent reputation. In order to validate our approach, we performed an experimental campaign by means of a simulated framework, which allowed us to verify that, by our approach, devices have not any economic convenience to performs misleading behaviors. Moreover, further experimental results have shown that our approach is able to detect the nature of the active agents in the systems (i.e., honest and malicious), with an accuracy of not less than 11% compared to the best competitor tested and highlighting a high resilience with respect to some malicious activities.

Struggling for legitimate meaning: Agent–structure dynamics in German filmmaking

Given the state sponsorship of film production in Germany, this article examines general mechanisms in the formation of meaning in German filmmaking. With reference to Schimank’s framework of agent–structure dynamics and based on a constructivist understanding of the world, the results of 97 expert interviews with screenwriters, directors, producers, distributors, cinema theatre operators, funding representatives and public television editors, as well as document analyses, show that the medium’s construction of reality is anything but unconditioned. On the contrary, due to the fundamental role of film funding and public television in the agent constellations intertwined with social structures that shape the film production process in Germany, the medium’s key communicators are confronted with expectations that go far beyond economic parameters. More precisely, the article reveals that German filmmaking reflects a political dimension, and expresses hierarchies and constraints that prompt struggles for legitimate meaning and challenge any autonomous practice in the field.