Agents

Adaptive control in the face of uncertainty involves making online predictions about events in order to plan actions to reliably achieve desirable outcomes. However, we often face situations in which the effects of our actions are probabilistic (e.g., returns on our investments), and the environment itself is dynamic, - it can change irrespective of any action we decide to take, and in a relatively stable or unstable way (e.g., economic climate). Therefore, the problem of control concerns learning to isolate the effects that are generated directly by our actions, from those that occur independently of them. We face this problem in a host of uncertain dynamic environments: ecological (using fertilizers to increase crop yield), economic (investing in real estate), industrial (operating nuclear power plants), mechanical (driving cars) and medical (stemming outbreaks of disease). Thus far there has been little attempt at a synthesis of the amassing research directed towards understanding control in the face of uncertainty. The aim of this book is to bring together theoretical and empirical research from disparate disciplines spanning Social, Cognitive and Clinical Psychology, Human Factors and Neuroscience to Engineering and Machine learning. It serves as a round up of the different techniques used to examine issues concerning control (prediction, agency, causality, uncertainty) and unifies a substantial amount of disparate research. The general framework (Monitoring and control- MC framework) used to achieve this proposes two central ideas (agency, uncertainty) that have important ramifications across disciplines. In sum, control processes involve self-regulatory mechanisms that enable us to make subjective estimates of success, which helps to anchor our interpretation of uncertain events. In addition, these estimates are integrated with estimations of rates of change of the events in the environment which helps us track the outcomes we aim to control.

""In this paper, I explore extensively the dynamics of patronage and execution of silverware in Rome after designs by Giulio Romano. By discussing a large corpus of mostly unpublished material, I highlight the role of intermediaries and agents in the commissioning process of silver objects, and show the difficulties that the Gonzaga agents encounterd in their dealing with goldsmiths. As well as managing considerable sums, Gonzaga agents had to be aware of issues of identity and decorum and pursue their goals with a high degree of independence. This meant that sometimes they had take important decision with regard to economic, aesthetic and formal issues and to negotiate their own position within the fiercely competitive Roman artistic world. Skill, trust and reputation were key-values in these dynamics.

I presented this material at:
- Giulio Romano’s Silverware in the Making. Authorship, Patronage and the Role of Agents, Renaissance Conference of Southern California (RCSCA.org). Panel organized by Prof. Sheryl E. Reiss. February 6, 2010
- Invenzione e committenza nella produzione di argenti di Giulio Romano. International conference: “Giulio Romano e l’arte del Cinquecento”, organised by Ugo Bazzotti, May 29, 2009.""

Review of P. Demeulenaere (ed.), Analytical Sociology and Social
Mechanisms. New York: Cambridge University Press, 2011.

Why should we introduce the notion of ‘analytical sociology’ into the field of sociology, and why should it be linked to the concept of ‘mechanism’? With these two principal questions, Pierre Demeulenaere, Professor of Sociological Theory and Philosophy of the Social Sciences at the University of Paris-Sorbonne, opens his Analytical Sociology and Social Mechanisms, a collection of thirteen papers written by social scientists and philosophers of the social sciences. Not every contributor should be considered an analytical sociologist. Rather than being a manifesto either pro or contra analytical sociology and the use of mechanisms, it is an attempt to reflect upon the key issues involved in sociological explanation. Even though several chapters raise very interesting points, the overall impression one gets from this book is that analytical sociology fails to redeem its main promise, viz. to add clarity, precision, and conceptual rigor to sociology, especially regarding one of its most central concepts: ‘mechanism’.

This paper proposes a Multiagent Based Multipath Routing (MBMR) using a set of static and mobile agents by employing localization technique. The operation of proposed routing technique can be briefly explained as follows. (1) Anchor nodes are deployed evenly over the network environment. (2) Unknown sensor nodes are deployed randomly over network environment and these nodes perform localization. (3) Source node computes the shortest route to destination node through arbitrary midpoint node and intermediate nodes. (4) Source node generates mobile agents for partial route discovery, which traverses to destination node through the midpoint and intermediate nodes by carrying information. (5) Mobile agents update the destination node with carried information. (6) Destination node computes route weight factor for all the routes discovered by mobile agents. (7) Destination node computes the node disjoint routes and it selects routes depending on the criticalness of event for communication. The performance of the proposed scheme is evaluated in terms of performance parameters such as localization error, network lifetime, energy consumption, cost factor, packet delivery ratio, and latency.

Today, mechanisms and mechanistic explanation are very popular in philosophy of science and are deemed a welcome alternative to laws of nature and deductive-nomological explanation. Starting from Mitchell’s pragmatic notion of laws, I cast doubt on their status as a genuine alternative. I argue that (1) all complex-systems mechanisms ontologically must rely on stable regularities, while (2) the reverse need not hold. Analogously, (3) models of mechanisms must incorporate pragmatic laws, while (4) such laws themselves need not always refer to underlying mechanisms. Finally, I show that Mitchell’s account is more encompassing than the mechanistic account.

Abstract. This paper presents the architecture of an Intelligent Learning Management System (ILMS) applied to Moodle. In particular, we present the design and implementation of an agent that select the teaching strategy according to the student's learning style. The chosen teaching strategy is used to filter the learning objects that are displayed to the students. Keywords: Intelligent Learning Management System, teaching strategy, learning style, personalization content.

A basic claim of this paper is that the foundational theoretical problem of the social sciences — the possibility of unconscious, unplanned forms of cooperation and intelligence among intentional agents (the very hard issue of the ‘invisible hand’, of the ‘spontaneous social order’ but also of ‘social functions’) — will eventually be clarified thanks to the contribution of AI (and, in particular, of cognitive Agent modelling, learning, and MAS) and its entering the social simulation domain. After introducing Multi-Agent-Based Social Simulation and its trends, the limits of the very popular notion of ‘emergence’ are discussed, Smith’s and Hayek’s view of ‘spontaneous social order’ are critically introduced, and serious contradictions in the theory of ‘social functions’ among intentional agents are pointed out. The problem is how to reconcile the ‘external’ teleology that orients the agent’s behaviour with the ‘internal’ teleology governing it. In order to account for the functional character of intentional action, we need a somewhat sophisticated model of intention, and a different view of layered cognitive architectures combining explicit beliefs and goals with association and conditioning. On such a basis we sketch a model of unknown functions impinging on intentional actions through a high level form of (MA) reinforcement learning. This model accounts for both eu-functions and dys-functions, autonomous and heteronomous functions. It is argued that, in order to reproduce some behaviour, its effects should not necessarily be ‘good’, i.e. useful for the goal of the agent or of some higher macro-system.  2001 Elsevier Science B.V. All rights reserved.

Today, complex-systems mechanisms and mechanistic explanation are very popular in philosophy of science and are deemed a welcome alternative to the decried laws of nature and to D-N explanation. However, starting from a relaxed, pragmatic notion of regularities and laws, I will cast doubt on their status as a genuine alternative. I will argue that (1) all complex-systems mechanisms ontologically must rely on stable regularities, while (2) it is not obvious that all such regularities must rely on an underlying mechanism. Analogously, (3) models of mechanisms must incorporate (and hence are epistemologically dependent on) pragmatic laws, while (4) such laws themselves needn’t always represent mechanisms. As valuable as the concepts of mechanism and mechanistic explanation are, they cannot replace regularities nor undermine their relevance regarding explanation.

In three experiments we investigated whether two procedures of acquiring knowledge about the same causal structure, predictive learning (from causes to effects) versus diagnostic learning (from effects to causes), would lead to different base rate use in diagnostic judgments. Results showed that learners are capable of incorporating base rate information in their judgments regardless of the direction in which the causal structure is learned. However, this only holds true for relatively simple scenarios. When complexity was increased, base rates were only used after diagnostic learning, but were largely neglected after predictive learning. It could be shown that this asymmetry is not due to a failure of encoding base rates in predictive learning because participants in all conditions were fairly good at reporting them. The findings present challenges for all theories of causal learning.

The use of rational agents for modelling real world problems has both been heavily investigated and become well accepted, with BDI (Beliefs, Desires, and Intentions) Logic being a widely used architecture to represent and rea-son about rational agency. However, in the real world, we often have to deal with different levels of confidence in the beliefs we hold, desires we have, and intentions that we commit to. This paper extends our previous frame-work that integrated qualitative levels of beliefs, desires, and intentions into BDI Logic. We describe an expanded set of axioms and properties of the extended logic. We present a modular structure for the semantics which in-volves a non-normal Kripke type semantics that may be used for other agent systems. Further, we demonstrate the usefulness of our framework with a scheduling task example.

The Agency Costs resulting from the conflict between principals and agents,
due to the separation of ownership and control in the capitalist enterprise,
was analyzed in many works of the last century. Could we assume that
the current globalization of companies has increased the differences
between principals and agents? Does this scenario lead to an increase
in Agency Costs in those companies with the greatest presence in the
global economy? The objective of this article is to compile studies that will
allow us to answer these questions affirmatively and conclude that large
transnational, multinational or global companies involve higher Agency
Costs than those that remain within national borders

The scientific analysis and management of fisheries has been dominated by three institutional orthodoxies: state regulation, market exchange and community-based management. This paper argues that these static, substantive, metaphors obscure a consideration of the dynamic temporal interplay between stakeholders and their environment. To address this vacuity the paper outlines a process-sociological approach and applies it to the historical development of fisheries management in the Tam Giang Lagoon, in Central Vietnam. The analysis shows that the interplay between different groups of fishers, the Vietnamese state, and the lagoon environment, over time, structures future struggles. As such, this interplay needs to be understood as a causal force of its own in the development of fisheries management. Endorsement of the analysis raises important questions concerning the potential of policy and planning to control fisheries.

The World Wide Web has unleashed people's imagination and a plethora of technologies have emerged. As these technologies have sprung up to address different requirements, it has become imperative to understand how these different technologies fit together. This paper presents the state of the art of these different technologies and tries to present a coherent vision of their inter-operation.

ConAg is a reusable framework, written in Java, for creating \conscious " software agents. The system provides an application skeleton that can be customized by developers. Its particular focus is on these agents ' \consciousness " mechanism. A \conscious " software agent is a cognitive agent that integrates numerous articial intelligence mechanisms to implement the global workspace theory, a psychological theory of mind. This article gives an overview of \conscious " software agents and it describes the architectural style of \conscious " software agents, from which ConAg is constructed.

This research work describes the setup of framework for testing the performance of intrinsically motivated swarm agents over extrinsic motivation. The performance is tested through the simulation. The result demonstrates that agents with intrinsic motivation for specific goal have high metacognitive ability. It also shows group performance of agents with metacognitive ability is better than the group of agents with extrinsic motivation exhibiting cognitive ability. Goal setting theory of motivation is applied to the group of agents in order to analyse the intelligent behaviour of the agents. This research is mainly focusing on why and how group performance by swarm agents is better than individuals. This approach requires design of ambient testbed where swarm agents demonstrate cognitive actions to metacognitive actions. This research is aiming to prove that group performance by swarm agents is higher due to type of agents chosen with intrinsic motivation and thus proves intrinsic motivation is better than extrinsic motivation. Agent behaviour in a group can be analysed using different metrics like resource collection, life expectancy, level of motivation and task assigned.

Abstract: With steadily evolving new paradigms for technology enhanced learning, educational word games such as quizzes, puzzles and quests raise new appeal and motivation for students in following game based educational processes. Traditional word games may be applied more successfully to game based learning in given scientific domain provided they are highly oriented to the content and problems of that domain. Such games may be more efficient if they include artificial agents simulating opponents, advisors or collaborators of the player. Authors present a semantic structuring model of learning content for logic and word games serving for educational purposes and, next, show the place of artificial agents within the game construction and possible ways of agent’s realization. There are given results from practical experiments with playing a memory game using the semantic content model, without and with agents integrated into the game.

In today’s increasingly competitive global market, most enterprises place great stress on reducing order fulfillment costs, minimizing time-to-market, and maximizing product quality. The desire of businesses to achieve these goals has seen a shift from a make-to-stock paradigm to a make-to-order paradigm. The success of the make-to-order paradigm requires robust and efficient solutions to the supply-chain-integration problem in the business-to-business (B2B) environment. Recent Internet-based solutions to this problem have enabled efficient and effective information sharing among trading partners (customers, manufacturers, and suppliers). Here we present an integration framework for integrating supply chain operations among such trading partners. A supply chain scenario is constructed, for which an integration framework is proposed based on by concept of business process specifications (BPS) and technology of distributed agents. We model the BPS, which specifies message choreographi...

Objec􀆟ves Vascular endothelial cells play an important role in the regula􀆟on of vascular vessels with
ac􀆟ve substances, such as vascular endothelial growth factor are responsible. The present study tried
to inves􀆟gate the effect of resistance training on the density of old women’s endothelial growth factor.
Methods & Materials Twenty-four postmenopausal women (age Mean±SD=67.37±6.02,
height=153.02±8.12, weight=65.78±12.03, BMI=26.87±4.16, fat=18.61±3.65, and WHR=0.92±0.4) were
purposefully chosen and randomly divided into control and experimental groups. Each group consisted
of 12 women. Experimental group did resistance training for eight weeks as follows: three sessions per
week with 40% to 65% intensity of a maximum repe􀆟􀆟on and 5% overload a􀅌er each 6 sessions. Before
and a􀅌er 8 weeks of training, res􀆟ng levels of vascular endothelial growth factor’s density was measured
and recorded. The normality of the data was determined by the Kolmogorov-Smirnov test. Paired
samples t test was run for within group varia􀆟ons and independent samples t test was run for between
groups’ varia􀆟ons. The significant level was lower than 0.05.
Results average and standard devia􀆟on of before and a􀅌er eight weeks concentra􀆟on of tour vascular
endothelial growth inn control group was respec􀆟vely 26.61±3.62, 25.72±2.88 and also in experimental
group was 23.81±2.18, 26.48±4.48, 23.81±2.18 respec􀆟vely; therefore the mean differences of vascular
endothelial growth factor was significant in pretest and pos􀆩est of experimental group (P=0.015) while
it wasn’t significant in control group P=0.862.
Conclusion The results showed the significant effect of resistance training on the reduc􀆟on of the density
of vascular endothelial growth factor of the old women’s plasma (P=0.019).

A framework for classifying children's explanations about dissolution, floating, magnetic forces, plant nutrition and the water cycle is presented. Explanations are classified as naturalistic, non-naturalisic, or synthetic.

Hybrid agent-landscape models are used as an environment in which to study humans, the environment, and their dynamics. To provide flexibility in model design, expressiveness, and modification, the environment models and human agent models should be developed independently. While retaining each model's individuality, the models can be composed to create a model of a complex, hybrid agent-landscape system. This should allow for a much more in-depth analysis of each model independently, as well as a study of their interactions. To create such a modeling environment requires a look beyond a simple interface between two models. It may require that the models' formalisms be composed, their execution be synchronized, their architectures be integrated, and a common visualization be created to provide a whole-system data view during simulation. This paper discusses the complexities of such an undertaking.

Our aim is to produce a focused crawler that, given one or a number of sample pages, will crawl to all similar pages on the web as efficiently as possible. A key problem in achieving this goal is assigning credit to documents along a crawl path, so that the system can learn which documents lead toward goal documents and can thus efficiently search in a best first manner. To address this problem, we construct an artificial economy of autonomous agents. Each agent buys and sells web pages and is compensated when it buys a goal page, when another agent buys the current set of uncrawled web pages, or when future agents buy a goal page. The economy serves to push money up from goal pages, compensating agents that buy useful pages. Inappropriate agents go broke and new agents are created, and the system evolves agents whose bids accurately estimate the utility of adding pages to the search. The system is found to outperform a Bayesian focused crawler in our experiments.

We present an approach for mapping an OWL ontology into Java. The basic idea is to create a set of Java interfaces and classes from an OWL ontology such that an instance of a Java class represents an instance of a single class of the ontology with most of its properties, classrelationships and restriction-definitions maintained. We note that there exist some fundamental semantic differences between Description Logic (DL) and Object Oriented (OO) systems, primarily related to completeness and satisfiability. We present various ways in which we aim to minimize the impact of such differences, and show how to map a large part of the much richer OWL semantics into Java. Finally, we sketch the HarmonIA framework, which is used for the automatic generation of agent systems from institution specifications, and whose OWL Ontology Creation module was the basis for the tool presented in this paper.

This paper presents a multiagent system to manage and monitor the surveillance routes of the security personnel in the construction sector. Several concepts, derived from the Artificial Intelligence, have been taken into account for the system design, obtaining context information in order to automatically monitor the surveillance routes in execution time. The system is composed of intelligent agents with reasoning and planning capabilities, as well as of a set of technologies allowing users to interact with the system and the environment in a natural, unobstructive and easy way.