Informes técnicos de investigación DSIC-TI-IA

Colección de documentos de trabajo, informes técnicos y otro material de investigación realizado por los investigadores del Departamento de Sistemas Informáticos y Computación (DSIC). Grupo Tecnologia Informatica - Inteligencia Artificial de la Universitat Politècnica de València en el desarrollo de sus investigaciones.

URI permanente para esta colecciónhttps://riunet.upv.es/handle/10251/8361

Examinar

Mostrando 1 - 16 de 16

Access status: Abierto ,
Designing an Affective Intelligent Agent on GenIA³
(Universitat Politècnica de València, 2017-03-06) Alfonso Espinosa, Bexy
GenIA³ is a General-purpose Intelligent Affective Agent Architecture, which can be committed with specific psychological theories to create the design of the final agent. Intelligent affective agents can be implemented by using the default design of GenIA³. Also GenIA³ helps experts on fields like psychology or behavioral computing, to provide more precise and refined ways of describing each particular affective process, facilitating the abstraction from irrelevant implementation or design details, and offering a default design for the main processes. Nevertheless an extensive set of domains need to be tested in order to properly validate and refine GenIA³. In this work we describe the default design of GenIA³, and we propose an alternative design which is based on a model of emotions previously proposed. This illustrates the flexibility of GenIA³ and may inspire other alternative designs.
Access status: Abierto ,
Toward a Systematic Development of Affective Intelligent Agents
(Universitat Politècnica de València, 2016-04-12) Alfonso Espinosa, Bexy; Vivancos, Emilio; Botti V.; Departamento de Sistemas Informáticos y Computación; Escuela Técnica Superior de Ingeniería Informática; Instituto Universitario Valenciano de Investigación en Inteligencia Artificial
The representation of the knowledge that is used for the specification of affective processes in agents, is almost as diverse as number of approaches that have addressed this issue. This diversity is due, to a large extent, to the need of systematic guidelines and standards that support computer scientists on the creation of affective models and architectures. Our aim is to perform a further step towards the standardization of this process, in order to improve the creation and enhancement of affective agent languages, architectures, and models. We offer a method to build affective BDI (Beliefs, Desires, and Intentions) agents, adapted to the problem to solve, and specifically, adapted to the way affect influences the agent behavior. To this end we offer GenIA³ , a General-purpose Intelligent Affective Agent Architecture, which can be committed with specific psychological theories to create the architecture of the final agent. We also offer general guidelines that allow to define the processes performed in the agent architecture. These guidelines allow to select and adapt a BDI agent platform in order to include the processes of the proposed agent architecture and adapt a BDI agent language to include the representation of the required affect-related attributes.
Access status: Abierto ,
Magentix 2 User's Manual
(Universitat Politècnica de València, 2015-03-30) Botti V.; Argente, Estefanía; Alemany Bordera, José; Bellver Faus, Joan; Búrdalo Rapa, Luis Antonio; Carrascosa Casamayor, Carlos; Criado Pacheco, Natalia; de la Fuente Anuarbe, Miguel Ángel; Del Val Noguera, Elena; Esparcia García, Sergio; Espinosa Minguet, Agustín Rafael; García-Fornes, A; Garcia Marques, Mª Emilia; Rarcia Pardo Gimenez de lo sGalanes, Juna Ángel; Giret Boggino, Adriana Susana; Hérnandez López, Luís; Heras, Stella; Jordán, Jaume; Jorge Cano, Javier; Julian, Vicente; López Fogués, Ricard; Mejías Rodríguez, José Manuel; Alberola Oltra, Juan Miguel; Such Oltra, José Miguel; Mulet Mengual, Luis; Navarro Llácer, Martín; Pajares Ferrando, Sergio; Palanca Cámara, Javier; Palomares Chust, Alberto; Pérez, Pedro; Rebollo Pedruelo, Miguel; Rodrigo Solaz, Mario; Ruíz, José Vicente; Soler Bayona, José Vicente; Terrasa Barrena, Andrés Martín; Valero Cubas, Soledad; Departamento de Sistemas Informáticos y Computación; Escuela Técnica Superior de Ingeniería Aeroespacial y Diseño Industrial; Escuela Politécnica Superior de Gandia; Escuela Técnica Superior de Ingeniería Informática; Instituto Universitario Valenciano de Investigación en Inteligencia Artificial
Magentix2 is an agent platform for open Multiagent Systems. Its main objective is to bring agent technology to real domains: business, industry, logistics, e-commerce, health-care, etc. Magentix2 platform is proposed as a continuation of the first Magentix platform. The final goal is to extend the functionalities of Magentix, providing new services and tools to allow the secure and optimized management of open Multiagent Systems. Nowadays, Magentix2 provides support at three levels: - Organization level, technologies and techniques related to agent societies. - Interaction level, technologies and techniques related to communications between agents. - Agent level, technologies and techniques related to individual agents (such as reasoningand learning). Thus, Magentix2 platform uses technologies with the necessary capacity to cope with the dynamism of the system topology and with flexible interactions, which are both natural consequences of the distributed and autonomous nature of its components. In this sense, the platform has been extended in order to support flexible interaction protocols and conversations, indirect communication and interactions among agent organizations. Moreover, other important aspects cover by the Magentix2 project are the security issues.
Access status: Abierto ,
EEG-based biomarkers and emotional response to the Ultimatum Game
(Universitat Politècnica de València, 2014-10-15) V. Botti; Y Gomez; V. Martinez; J. Vila
This paper presents a basic neuro-experiment to explore the possibility of the application of EEG-based biomarkers to analyze emotional response to the ultimatum game. Using a basic EEG registration tool (Emotiv EPOC technology), the paper presents two EEG-based biomarkers (F8-EP and ALPHA8-AP) obtained from an evoked potential and accumulated power approaches, which seem to be related with the responder’s emotions when facing an unfair offer in the Ultimatum Game.
Access status: Abierto ,
Extending a BDI agents' architecture with open emotional components
(Universitat Politècnica de València, 2014-07) Alfonso Espinosa, Bexy; Vivancos, Emilio; Botti V.; Departamento de Sistemas Informáticos y Computación; Escuela Técnica Superior de Ingeniería Informática; Instituto Universitario Valenciano de Investigación en Inteligencia Artificial
Recently an increasing amount of research focuses on improving agents believability by adding affective features to its traditional modeling. This is probably due to the demands of reaching ever more realistic behaviors on agents simulations which extends to several and diverse applications fields. The present work proposes O3A: an Open Affective Agent Architecture, which extends a traditional BDI agent architecture improving a practical reasoning with more “human” characteristics. This architecture tries to address disperse definitions combining the main elements of supporting psychological and neurological theories.
Access status: Abierto ,
mWater prototype review
(Universitat Politècnica de València, 2013-09-17) Botti V.; Garrido, Antonio; Giret Boggino, Adriana Susana; Noriega, Pablo; Departamento de Sistemas Informáticos y Computación; Escuela Técnica Superior de Ingeniería Aeroespacial y Diseño Industrial; Escuela Técnica Superior de Ingeniería Informática; Instituto Universitario Valenciano de Investigación en Inteligencia Artificial
This document reviews our current water policy-making decision-support framework, build on top of a regulated open Multi-Agent System (MAS),mWater [BGG+10, GGG+11], that models a flexible water-rights market. Our simulator focuses on the effect of regulations on demand and thus provides means to explore the interplay of norms and conventions that regulate trading (like trader eligibility conditions, tradeable features of rights, trading periods and price-fixing conventions), the assumptions about agent behaviour (individual preferences and risk attitude, or population profile mixtures) and market scenarios (water availability and use restrictions). A policy-maker would then assess the effects of those interactions by observing the evolution of the performance indicators (efficiency of use, price dynamics, welfare functions) (s)he designs. 1.2 Our
Access status: Abierto ,
mWater prototype #2 analysis and design
(Universitat Politècnica de València, 2013-09-17) Botti V.; Criado Pacheco, Natalia; Garrido, Antonio; Gimeno, J.A.; Giret Boggino, Adriana Susana; Noriega, Pablo; Departamento de Sistemas Informáticos y Computación; Escuela Técnica Superior de Ingeniería Aeroespacial y Diseño Industrial; Escuela Técnica Superior de Ingeniería Informática; Instituto Universitario Valenciano de Investigación en Inteligencia Artificial
mWater is a regulated open MAS that uses intelligent agents to manage a flexible water-right market. One of the main goals of mWater is to be used as a simulator to assist in decision-taking processes for policy makers. Our simulator focuses on demands and, in particular, on the type of regulatory (in terms of norms selection and agents behaviour), and market mechanisms that
Access status: Abierto ,
mWater prototype #3 analysis and design
(Universitat Politècnica de València, 2013-09-17) Botti V.; Garrido, Antonio; Giret Boggino, Adriana Susana; Noriega, Pablo; Bexi, Alfonso; Departamento de Sistemas Informáticos y Computación; Escuela Técnica Superior de Ingeniería Aeroespacial y Diseño Industrial; Escuela Técnica Superior de Ingeniería Informática; Instituto Universitario Valenciano de Investigación en Inteligencia Artificial
In themWater case study prototype #3 it has been used Magentix2 [1, 24, 3, 22, 4, 17] (for more details on Magentix2 see WP7 Deliverables) as the MAS platform for supporting the execution of the MAS system. The platform follows the FIPA standards [14] offering a set of useful mechanisms for the agents to communicate and also tools to allow programming agents in a high level language based on the BDI model. Magentix2 is an open system which facilitates the interaction between heterogeneous agents through FIPA-ACL messages. Also complex interactions can be carried out in a flexible an open way as conversations. The platform offers special structures to allow to use such conversations by considering a set of issues: In each conversation there are always two roles involved: Initiator and Participant. The first is the one who initiates the conversation, and the rest of agents play the Participant role. The conversation can be seen as a direct graph where nodes represent the actions to perform in each step of the conversation and arcs represent the transition between such states. Those steps allow to perform some actions and they can be of different kinds, for example: Begin, Final, Wait, Send, Receive, Action, etc. Conversations have a unique identifier that allows to manage them individually. 1
Access status: Abierto ,
mWater Prototype 3
(Universitat Politècnica de València, 2013-09-17) Botti V.; Garrido, Antonio; Giret Boggino, Adriana Susana; Noriega, P.; Gimeno, J.; Departamento de Sistemas Informáticos y Computación; Escuela Técnica Superior de Ingeniería Aeroespacial y Diseño Industrial; Escuela Técnica Superior de Ingeniería Informática; Instituto Universitario Valenciano de Investigación en Inteligencia Artificial
This report concerns the application of a regulated open Multi-Agent System (MAS), mWater, that uses intelligent agents to simulate a flexible water-right market. Our simulator focuses on demands and, in particular, on the type of regulatory (in terms of norms selection and agents behaviour), and market mechanisms that foster an efficient use of water while also trying to prevent conflicts among parties. In this scenario, a MAS plays a vital role as it allows us to define different norms, agents behaviour and roles, and assess their impact in the market, thus enhancing the quality and applicability of its results as a decision support tool.
Access status: Abierto ,
mWater Analysis
(Universitat Politècnica de València, 2013-09-17) Botti V.; Garrido, Antonio; Giret Boggino, Adriana Susana; Igual, Francesc; Noriega, Pablo; Igual; Departamento de Sistemas Informáticos y Computación; Escuela Técnica Superior de Ingeniería Aeroespacial y Diseño Industrial; Escuela Técnica Superior de Ingeniería Informática; Instituto Universitario Valenciano de Investigación en Inteligencia Artificial
The mWater scenario requires the expression and use of regulations of different sorts: from actual laws and regulations issued by governments, to policies and local regulations issued by basin managers, to social norms that prevail in a given community of users. Some will be regimented as part of the electronic institutional framework specification, but others need to be expressed in a declarative form so that one may reason about them, both off- and on-line, both at design and at run time, and both from the institutional (or legislative) perspective and the agent's individual perspective. Issues that are relevant in this respect range from the choice of expressive formalism to the decision-making strategies that agents might use to comply or disobey regulations. Thus, structural aspects like governance, dynamics of norms (also from the legislative and execution perspectives) as well as criteria to evaluate the effectiveness of norms may and need to be explored in the demonstrator.
Access status: Abierto ,
mWater prototype #3 review
(Universitat Politècnica de València, 2013-09-06) Garrido, Antonio; Botti V.; Giret Boggino, Adriana Susana; Alfonso Espinosa, Bexy; Noriega, Pablo; Departamento de Sistemas Informáticos y Computación; Escuela Técnica Superior de Ingeniería Aeroespacial y Diseño Industrial; Escuela Técnica Superior de Ingeniería Informática; Instituto Universitario Valenciano de Investigación en Inteligencia Artificial
mWater is a software demonstrator developed in the Agreement Technologies Project. It is a Multi-Agent System (MAS) application that implements a market for water rights, including the model and simulation of the water-right market itself, the basin, users, protocols, norms and grievance situations. mWater is motivated due to the fact that water scarcity is becoming a major concern in most countries, not only because it threatens the economic viability of current agricultural practices, but because it is likely to alter an already precarious balance among its different types of use.
Access status: Abierto ,
Domain Independent Temporal Planning in a Planning-Graph-Based Approach
(Universitat Politècnica de València, 2012-12-14) Onaindia De La Rivaherrera, Eva; Garrido, Antonio; Departamento de Sistemas Informáticos y Computación; Escuela Técnica Superior de Ingeniería Informática; Instituto Universitario Valenciano de Investigación en Inteligencia Artificial
Many planning domains have to deal with temporal features that can be expressed using durations that are associated to actions. Unfortunately, the conservative model of actions used in many existing temporal planners is not adequate for domains which require more expressive models. This paper presents a temporal planning approach that combines the principles of Graphplan and TGP and uses the information calculated in the planning graph to deal with a non-conservative model of actions that include local conditions and e ects. In this approach, we propose two strategies for search. The rst one is based on the Graphplan backward search. The second one is based on a least-commitment and heuristic search, and it attempts to overcome the main limitations of a chronological backtracking search when dealing with large temporal problems. This search has proved to be bene cial in the scalability of the planner and the experiments show that a planner using this new search is competitive with other state-of-the-art planners w.r.t. the plan quality
Access status: Abierto ,
Case-Based Argumentation Framework. Dialogue Protocol
(Universitat Politècnica de València, 2011-06-24) Heras, Stella; Botti V.; Julian, Vicente; Departamento de Sistemas Informáticos y Computación; Escuela Técnica Superior de Ingeniería Informática; Instituto Universitario Valenciano de Investigación en Inteligencia Artificial
On top of the simpler ability to interact, open MAS must include mechanisms for their agents to reach agreements by taking into account their social context. Argumentation provides MAS with a framework that assures a rational communication, which allows agents to reach agreements when conflicts of opinion arise. In this report we present the communication protocol that agents of a case-based argumentation framework use to interact when they engage in argumentation dialogues. The syntax and semantics of the framework are formalised and discussed.
Access status: Abierto ,
Case-Based Argumentation Framework. Strategies
(Universitat Politècnica de València, 2011-06-24) Heras, Stella; Botti V.; Julian, Vicente; Departamento de Sistemas Informáticos y Computación; Escuela Técnica Superior de Ingeniería Informática; Instituto Universitario Valenciano de Investigación en Inteligencia Artificial
In agent societies, agents perform complex tasks that require different levels of intelligence and give rise to interactions among them. From these interactions, conflicts of opinion can arise, specially when MAS become adaptive and open with heterogeneous agents dynamically entering in or leaving the system. Therefore, software agents willing to participate in this type of systems will require to include extra capabilities to explicitly represent and generate agreements on top of the simpler ability to interact. In addition, agents can take advantage of previous argumentation experiences to follow dialogue strategies and easily persuade other agents to accept their opinions. Our insight is that CBR can be very useful to manage argumentation in open MAS and devise argumentation strategies based on previous argumentation experiences. To demonstrate the foundations of this suggestion, this report presents the work that we have done to develop case-based argumentation strategies in agent societies. Thus, we propose a case-based argumentation framework for agent societies and define heuristic dialogue strategies based on it. The framework has been implemented and evaluated in a real customer support application.
Access status: Abierto ,
Case-Based Argumentation Framework. Reasoning Process
(Universitat Politècnica de València, 2011-06-24) Heras, Stella; Botti V.; Julian, Vicente; Departamento de Sistemas Informáticos y Computación; Escuela Técnica Superior de Ingeniería Informática; Instituto Universitario Valenciano de Investigación en Inteligencia Artificial
The capability of reaching agreements is a necessary feature that large computer systems where agents interoperate must include. In these systems, agents represent self-motivated entities that have a social context, including dependency relations among them, and different preferences and beliefs. Without agreement there is no cooperation and thus, complex tasks which require the interaction of agents with different points of view cannot be performed. In this work, we follow a case-based argumentation approach for the design and implementation of Multi-Agent Systems where agents reach agreements by arguing and improve their argumentation skills from experience. A set of knowledge resources and a reasoning process that agents can use to manage their positions and arguments are presented.
Access status: Abierto ,
A Computational Argumentation Framework for Agent Societies
(Universitat Politècnica de València, 2011-06-17) Heras, Stella; Botti V.; Julian, Vicente; Departamento de Sistemas Informáticos y Computación; Escuela Técnica Superior de Ingeniería Informática; Instituto Universitario Valenciano de Investigación en Inteligencia Artificial
Starting from the idea that the social context of agents determines the way in which agents can argue and reach agreements, this context should have a decisive influence in the computational representation of arguments. In this report, we advance research in the area of computational frameworks for agent argumentation by proposing a new argumentation framework (AF) for the design of open MAS in which the participating software agents are able to manage and exchange arguments between themselves taking into account the agents¿ social context. In order to do this, we have analysed the necessary requirements for this type of framework 1 and taken into account them in the design of our framework. Also, the knowledge resources that the agents can use to manage arguments in this framework are presented in this work. In addition, if heterogeneous agents can interact in the framework, they need a common language to represent arguments and argumentation processes. To cope with this, we have also designed an argumentation ontology to represent arguments and argumentation concepts in our framework.

Examinar

Envíos recientes