Agent-based Learning Research Articles

Agent-based learning of materials datasets from the scientific literature

An AI Agent for autonomous development of materials dataset from scientific literature.

Agent-Based Learning Model for the Obesity Paradox in RCC.

A recent study on the immunotherapy treatment of renal cell carcinoma reveals better outcomes in obese patients compared to lean subjects. This enigmatic contradiction has been explained, in the context of the debated obesity paradox, as the effect produced by the cell-cell interaction network on the tumor microenvironment during the immune response. To better understand this hypothesis, we provide a computational framework for the in silico study of the tumor behavior. The starting model of the tumor, based on the cell-cell interaction network, has been described as a multiagent system, whose simulation generates the hypothesized effects on the tumor microenvironment. The medical needs in the immunotherapy design meet the capabilities of a multiagent simulator to reproduce the dynamics of the cell-cell interaction network, meaning a reaction to environmental changes introduced through the experimental data.

A survey on agent learning architecture that adopts internet of things and wireless sensor networks

Trustworthy and reliable applications built using intelligent software agents aim to provide improved performance using its characteristics. Agents introduced in various architectures represent its functionality as functional elements of the architecture and shows the interaction between other components present in the architecture. The Internet of things (IoT) reveals as a frequent technology that allows accessing the physical objects present in the world. IoT systems utilize wireless sensor network to transmit and receive data by establishing communication. Wireless Sensor Networks transmits digital signals to the cyber-world for analyzing and processing the information into useful data by either formulating or communicating with the intelligent and innovative system. While talking about IoT and WSN, agents introduced in such environments assist in making decisions quickly by perceiving the input from the environment. The number of agents needed for an application depends upon the complexity of the problem. Multi-Agent architectures discussed in the article describe their association, roles, functionality and interaction. This paper gives a detailed survey of various agent/multi-agent learning architectures introduced over IoT and WSN. Moreover, this survey with the performance and the SWOT analysis on the Agent-based learning architecture helps the reader and paves a way to pursue research on Agent-based architectural deployment over IoT and WSN paradigms.

On the convergence of projective-simulation\u2013based reinforcement learning in Markov decision processes

In recent years, the interest in leveraging quantum effects for enhancing machine learning tasks has significantly increased. Many algorithms speeding up supervised and unsupervised learning were established. The first framework in which ways to exploit quantum resources specifically for the broader context of reinforcement learning were found is projective simulation. Projective simulation presents an agent-based reinforcement learning approach designed in a manner which may support quantum walk-based speedups. Although classical variants of projective simulation have been benchmarked against common reinforcement learning algorithms, very few formal theoretical analyses have been provided for its performance in standard learning scenarios. In this paper, we provide a detailed formal discussion of the properties of this model. Specifically, we prove that one version of the projective simulation model, understood as a reinforcement learning approach, converges to optimal behavior in a large class of Markov decision processes. This proof shows that a physically inspired approach to reinforcement learning can guarantee to converge.

A review of machine learning applications in wildfire science and management

Artificial intelligence has been applied in wildfire science and management since the 1990s, with early applications including neural networks and expert systems. Since then, the field has rapidly progressed congruently with the wide adoption of machine learning (ML) methods in the environmental sciences. Here, we present a scoping review of ML applications in wildfire science and management. Our overall objective is to improve awareness of ML methods among wildfire researchers and managers, as well as illustrate the diverse and challenging range of problems in wildfire science available to ML data scientists. To that end, we first present an overview of popular ML approaches used in wildfire science to date and then review the use of ML in wildfire science as broadly categorized into six problem domains, including (i) fuels characterization, fire detection, and mapping; (ii) fire weather and climate change; (iii) fire occurrence, susceptibility, and risk; (iv) fire behavior prediction; (v) fire effects; and (vi) fire management. Furthermore, we discuss the advantages and limitations of various ML approaches relating to data size, computational requirements, generalizability, and interpretability, as well as identify opportunities for future advances in the science and management of wildfires within a data science context. In total, to the end of 2019, we identified 300 relevant publications in which the most frequently used ML methods across problem domains included random forests, MaxEnt, artificial neural networks, decision trees, support vector machines, and genetic algorithms. As such, there exists opportunities to apply more current ML methods — including deep learning and agent-based learning — in the wildfire sciences, especially in instances involving very large multivariate datasets. We must recognize, however, that despite the ability of ML models to learn on their own, expertise in wildfire science is necessary to ensure realistic modelling of fire processes across multiple scales, while the complexity of some ML methods such as deep learning requires a dedicated and sophisticated knowledge of their application. Finally, we stress that the wildfire research and management communities play an active role in providing relevant, high-quality, and freely available wildfire data for use by practitioners of ML methods.

Predictive, Agent-Based, and Causal Machine Learning Models of U.S. Congressional Elections

This paper uses district-level data to highlight similarities and differences of a variety of U.S. congressional election models ranging from predictive statistical learning models to causal structural models and new causal statistical models. First, several statistical learning models are produced with high out-of-sample accuracy. Second, the causal impacts of candidate spending, incumbency, and voter registration are estimated using a structural agent-based model as in Kretschman and Mastronardi (2010). Finally, a new environment for election modeling – machine learning for causal inference – is introduced. A demonstration is provided in the form of a “causal forest,” which uses machine learning to produce accurate predictions and estimate heterogeneous treatment effects. The causal impact estimates from the agent-based and causal forest models are used to develop counterfactual arguments describing the efficiency of allocated candidate spending across districts. The results from this research are valuable to many stakeholders, including candidates, party committees, and voters.

Connectionist agent-based learning in bank-run decision making.

It is of utter importance for the policy makers, bankers, and investors to thoroughly understand the probability of bank-run (PBR) which was often neglected in the classical models. Bank-run is not merely due to miscoordination (Diamond and Dybvig, 1983) or deterioration of bank assets (Allen and Gale, 1998) but various factors. This paper presents the simulation results of the nonlinear dynamic probabilities of bank runs based on the global games approach, with the distinct assumption that heterogenous agents hold highly correlated but unidentical beliefs about the true payoffs. The specific technique used in the simulation is to let agents have an integrated cognitive-affective network. It is observed that, even when the economy is good, agents are significantly affected by the cognitive-affective network to react to bad news which might lead to bank-run. Both the rise of the late payoffs, R, and the early payoffs, r, will decrease the effect of the affective process. The increased risk sharing might or might not increase PBR, and the increase in late payoff is beneficial for preventing the bank run. This paper is one of the pioneers that links agent-based computational economics and behavioral economics.

Analytical sociology and computational social science

Analytical sociology focuses on social interactions among individuals and the hard-to-predict aggregate outcomes they bring about. It seeks to identify generalizable mechanisms giving rise to emergent properties of social systems which, in turn, feed back on individual decision-making. This research program benefits from computational tools such as agent-based simulations, machine learning, and large-scale web experiments, and has considerable overlap with the nascent field of computational social science. By providing relevant analytical tools to rigorously address sociology’s core questions, computational social science has the potential to advance sociology in a similar way that the introduction of econometrics advanced economics during the last half century. Computational social scientists from computer science and physics often see as their main task to establish empirical regularities which they view as “social laws.” From the perspective of the social sciences, references to social laws appear unfounded and misplaced, however, and in this article we outline how analytical sociology, with its theory-grounded approach to computational social science, can help to move the field forward from mere descriptions and predictions to the explanation of social phenomena.

Intelligent dynamic spectrum access using hybrid genetic operators

Abstract This paper presents a novel hybrid dynamic spectrum access approach, combining classical and stochastic flavors being augmented with new genetic operators, for multi-channel single-radio cognitive radio networks. Existing classical and stochastic approaches exhibit different advantages and disadvantages depending on network architecture. Our proposed approach exploits a delicate balance between the two different approaches for extracting advantages from both of them while limiting their disadvantages. Additionally, in our proposed approach, we boost up extent of the exploitation through designing new genetic operators. Furthermore, we provide a thorough performance evaluation of our approach using a widely used discrete event simulator called ns-2 . Here, we also simulate several existing approaches that are based on graph theory, game theory, heuristic, genetic algorithm, agent-based learning, and online learning. Simulation results demonstrate significant performance improvement using our proposed intelligent dynamic spectrum access approach over these state-of-the-art ones based on various standard metrics.

Examining the predictive relationship between personality and emotion traits and students’ agent-directed emotions: towards emotionally-adaptive agent-based learning environments

The current study examined the relationships between learners' ($$N = 123$$N=123) personality traits, the emotions they typically experience while studying (trait studying emotions), and the emotions they reported experiencing as a result of interacting with four pedagogical agents (agent-directed emotions) in MetaTutor, an advanced multi-agent learning environment. Overall, significant relationships between a subset of trait emotions (trait anger, trait anxiety) and personality traits (agreeableness, conscientiousness, and neuroticism) were found for four agent-directed emotions (enjoyment, pride, boredom, and neutral) though the relationships differed between pedagogical agents. These results demonstrate that some trait emotions and personality traits can be used to predict learners' emotions directed toward specific pedagogical agents (with different roles). Results provide suggestions for adapting pedagogical agents to support learners' (with certain characteristics; e.g., high in neuroticism or agreeableness) experience of adaptive emotions (e.g., enjoyment) and minimize their experience on non-adaptive emotions (e.g., boredom). Such an approach presents a scalable and easily implementable method for creating emotionally-adaptive, agent-based learning environments, and improving learner-pedagogical agent interactions in order to support learning.

Agent-Based RBF Network Classifier with Feature Selection in a Kernel Space

ABSTRACTThe article addresses the problem of feature selection in a kernel space and proposes the approach to selecting the most informative features for classification carried out using the RBF neural classifier. In the article, class-dependent and cluster-dependent methods for feature selection are considered. The process of feature selection is supported by the rotation-based ensemble technique. The feature selection problem is viewed as an optimization task solved by the agent-based population learning algorithm applied at the RBFN’s initialization and training stage. The proposed approach is validated experimentally, and the obtained results are compared with the results produced using other methods. Experiment results show that the proposed method of feature selection in a kernel space of the RBF neural networks can be considered as a useful approach to constructing high-quality RBFN-based classifiers.

Enhancing middle school students' scientific learning and motivation through agent‐based learning

AbstractFacing students' decreasing motivation to pursue scientific study, schools and educators need to coordinate new technologies with pedagogical agents to effectively sustain or promote students' scientific learning and motivation to learn. Although the provision of pedagogical agents in student learning has been studied previously, it is not clear what benefits the strategy might offer with regard to student motivation. This study proposes an agent‐based mechanism that integrates problem‐solving and inquiry‐based instructions to help students better understand complex scientific concepts and to sustain their motivation to learn science. In this study, a quasi‐experiment was conducted to evaluate the performance and feasibility of our proposed mechanism. The results revealed that the agent‐based mechanism was effective and feasible for enhancing students' learning and motivation to learn. The mechanism was associated with increases in the acquisition of knowledge when compared with the control group. Its effect in promoting and sustaining students' motivation was also statistically significant. Detailed discussions of the findings are provided in this study.

Toward a Feature-Driven Understanding of Students' Emotions during Interactions with Agent-Based Learning Environments

This selective review synthesizes and draws recommendations from the fields of affective computing, intelligent tutoring systems, and psychology to describe and discuss the emotions that learners report experiencing while interacting with agent-based learning environments (ABLEs). Theoretically driven explanations are provided that describe the relative effectiveness and ineffectiveness of different ABLE features to foster adaptive emotions (e.g., engagement, curiosity) vs. non-adaptive emotions (e.g., frustration, boredom) in six different environments. This review provides an analytical lens to evaluate and improve upon research with ABLEs by identifying specific system features and their relationship with learners' appraisals and emotions.

Market Integration of Distributed Resources Through Coordinated Frequency and Price Droop

Given the appropriate control strategies and market rules, distributed energy resources (DER) organized into microgrids have the potential to be active participants in regional energy and ancillary services markets. This paper integrates a price-based control mechanism into the conventional frequency droop control mechanism to present an integrated droop-based control framework that could facilitate greater DER penetration in the power system and electricity markets. The proposed control framework makes use of the concept of “price droop,” and introduces the idea of a “system frequency elasticity”. These two concepts combine the power system control concept of frequency droop and the economic concept of own price elasticity. The framework consists of a centralized agent-based learning process and a decentralized operational response. The applicability of the proposed control mechanism is demonstrated through a series of case studies. Results highlight the need to consider the dynamics of the existing system in the design of new technical and market strategies to allow for the integration of DER and microgrids into the power system.

Identification of Motion-Based Action Potentials in Neural Bundles

Realizing a nerve signal based prostheses control or limb stimulation is a great challenge in medical technology. It requires a recording and an identification process of the motion-based action potentials of motor and sensory nerves within the corresponding neural bundle. Two additional key factors are used by multi agent-based learning algorithm: The anatomical disposition of the nerves within the neural bundle and the inverse kinematic. In this paper the authors introduce the Smart Modular Biosignal Acquisition, Identification and Control System and its application environment. They present the different process levels and their characteristic identification contribution and they give an overview of the multi-agent based identification framework. The authors show the verification environment and present results regarding the first-level identification procedure.

Individual Syllabus for Personalized Learner-Centric E-Courses in E-Learning and M-Learning

Most of e-learning and m-learning systems are course-centric. These systems provided services that concentrated on course material and pedagogical. They did not take into account varieties of student levels, skills, interests or preferences. This paper provides a design of an approach for personalized and self-adapted agent-based learning systems for enhancing e-learning and mobile learning (m-learning) services to be learner-centric. It presents a modeling of goals of different learners of a corporate training in computer courses in an educational institute. It figures how to customize and personalize learning paths (course syllabus) for e-learning and m-learning platforms. The delivering of e-courses become personalized learner-centric, which improves learning outcome, satisfaction of learners and enhances education.

Training Example Generation Method for Supervised Learning Agents in Sequential Scenarios

In this paper we propose a method of training example generation from agent's experience, which is suitable for sequential sce- narios. The experience consists of the agent's observations and its action records. Examples generated are used by the agent to learn a classifier, which is used to make decisions about its strategy in the following problem instances. The method is tested in a Sovereign environment, which is an economics simulation created to test agent-based learning. Experimental results show that an agent using the proposed methods is able to learn and achieves better results than random and heuristic agents.

Gendered socialization with an embodied agent: Creating a social and affable mathematics learning environment for middle-grade females.

This study examined whether or not embodied-agent-based learning would help middle-grade females have more positive mathematics learning experiences. The study used an explanatory mixed methods research design. First, a classroom-based experiment was conducted with one hundred twenty 9th graders learning introductory algebra (53% male and 47% female; 51% Caucasian and 49% Latino). The results revealed that learner gender was a significant factor in the learners’ evaluations of their agent (η² = .07), the learners’ task-specific attitudes (η² = .05), and their task-specific self-efficacy (η² = .06). In-depth interviews were then conducted with 22 students selected from the experiment participants. The interviews revealed that Latina and Caucasian females built a different type of relationship with their agent and reported more positive learning experiences as compared with Caucasian males. The females’ favorable view of the agent-based learning was largely influenced by their everyday classroom experiences, implying that students’ learning experience in real and virtual spaces was interconnected.

AGENT-BASED APPROACH TO THE DESIGN OF RBF NETWORKS

This article proposes a novel approach to the radial basis function network (RBFN) design. Its main idea is to apply the agent-based population learning algorithm to the task of initialization and training RBFNs. The approach allows for an effective network initialization and estimation of its output weights. The initialization involves two stages, where in the first one initial clusters are produced using the similarity-based procedure and next, in the second stage, prototypes (centroids) from the thus-obtained clusters are selected. The agent-based population learning algorithm is used to select prototypes. In the proposed implementation of the algorithm, both tasks—RBFN initialization and RBFN training—are carried out by a team of agents executing various local search procedures and cooperating with a view to determine the solution to the RBFN design problem at hand. The performance of the RBFN constructed using the proposed agent-based approach is analyzed and evaluated. The proposed approach is also compared with different RBFN initialization and training procedures in the literature.

SELECTING A REPRESENTATIVE DATA SET OF THE REQUIRED SIZE USING THE AGENT-BASED POPULATION LEARNING ALGORITHM

The aim of this article is to propose and evaluate an agent-based population learning algorithm generating, through prototype selection, a representative training data set of the required size. The main feature of the proposed approach is selection of the representative feature vectors, called prototypes, from clusters of feature vectors constructed over the original training data set under the assumption that from each cluster a single prototype is obtained. Thus, the number of clusters produced from the original training data set has a direct influence on the size of the reduced data set. The process of selection is executed by a team of agents, which execute various local search procedures and cooperate to determine a solution to the instance reduction problem, aiming at obtaining a compact representation of the data set. Rules for agent cooperation during the clustering and selection processes are defined within the so-called working strategy used by the team of agents (A-Team) in question. The article proposes a set of procedures that are used by agents to produce clusters and select prototypes. The approach is validated experimentally using well-known benchmark data sets. In addition to the computational experiment used to validate the model, the article investigates the efficiency and performance of two different working strategies used by the proposed A-Team. Because the proposed approach is based on the population learning algorithm, which belongs to the class of the population-based methods, an evaluation of the influence of the population of solution size on the performance of the algorithm is also included.