Skip to main content
Fachgebiete
Automobil + Motoren Bauwesen + Immobilien Business IT + Informatik Elektrotechnik + Elektronik Energie + Nachhaltigkeit Finance + Banking Management + Führung Marketing + Vertrieb Maschinenbau + Werkstoffe Versicherung + Risiko
DE
EN
Themenseiten
Organisationspsychologie Projektmanagement Marketing Smart Manufacturing
Bücher
Zeitschriften
Weitere Formate
Podcasts Webinare Technik Webinare Wirtschaft Kongresse Veranstaltungskalender Awards
Jetzt Einzelzugang starten
Zugang für Unternehmen
Referenzkunden
MTZ-Fachkongress

Antriebe und Energiesysteme von morgen


Austausch von Automobil- und Energiewirtschaft

Am 14./15. Mai geht es in Chemnitz um die Mobilitätswende durch Elektro- und Wasserstofffahrzeuge.
Erweiterte Suche
Anmelden
nach oben

2009 | Buch

The Prediction of Partners’ Behaviors in Self-interested Agents Kapitel lesen Erstes Kapitel lesen

Advances in Agent-Based Complex Automated Negotiations

herausgegeben von: Takayuki Ito, Minjie Zhang, Valentin Robu, Shaheen Fatima, Tokuro Matsuo

Verlag: Springer Berlin Heidelberg

Buchreihe : Studies in Computational Intelligence

Enthalten in: Springer Professional "Wirtschaft+Technik" , Springer Professional "Technik" , Springer Professional "Wirtschaft"

Einloggen, um Zugang zu erhalten
insite
SUCHEN

Über dieses Buch

Complex Automated Negotiations have been widely studied and are becoming an important, emerging area in the field of Autonomous Agents and Multi-Agent Systems. In general, automated negotiations can be complex, since there are a lot of factors that characterize such negotiations. These factors include the number of issues, dependency between issues, representation of utility, negotiation protocol, negotiation form (bilateral or multi-party), time constraints, etc. Software agents can support automation or simulation of such complex negotiations on the behalf of their owners, and can provide them with adequate bargaining strategies. In many multi-issue bargaining settings, negotiation becomes more than a zero-sum game, so bargaining agents have an incentive to cooperate in order to achieve efficient win-win agreements. Also, in a complex negotiation, there could be multiple issues that are interdependent. Thus, agent’s utility will become more complex than simple utility functions. Further, negotiation forms and protocols could be different between bilateral situations and multi-party situations. To realize such a complex automated negotiati on, we have to incorporate advanced Artificial Intelligence technologies includes search, CSP, graphical utility models, Bays nets, auctions, utility graphs, predicting and learning methods. Applications could include e-commerce tools, decisionmaking support tools, negotiation support tools, collaboration tools, etc.

These issues are explored by researchers from different communities in Autonomous Agents and Multi-Agent systems. They are, for instance, being studied in agent negotiation, multi-issue negotiations, auctions, mechanism design, electronic commerce, voting, secure protocols, matchmaking & brokering, argumentation, and co-operation mechanisms. This book is also edited from some aspects of negotiation researches including theoretical mechanism design of trading based on auctions, allocation mechanism based on negotiation among multi-agent, case-study and analysis of automated negotiations, data engineering issues in negotiations, and so on.

Inhaltsverzeichnis

Frontmatter
The Prediction of Partners’ Behaviors in Self-interested Agents
Summary
Multi-issue negotiation protocols represent a promising field since most negotiation problems in the real world involve multiple issues. Our work focuses on negotiation with interdependent issues, in which agent utility functions are nonlinear. We have proposed the multi-round representative-based protocol that utilizes the amount of agents’ private information revealed. However, the detailed effect of the representative selection method has not been shown. In this paper, we investigate the effect of the revealed area based selection method (RAS) in which agents who revealed larger utility area will be selected representatives. In the experiments, we compare the selection method with the random selection method in which representative agents are randomly selected. As a result, RAS is better for the optimality and the failure rate of finding solutions. Also, the fairness of the number of times to be representative agents is better. Moreover, we demonstrate its effect to the optimality and the failure rate in the experiment.
Fenghui Ren, Minjie Zhang
Sequential Auctions for Common Value Objects with Budget Constrained Bidders
Summary
This paper analyzes sequential auctions for budget constrained bidders, for multiple heterogeneous common value objects. In such auctions, the bidders’ problem is to determine how much to bid in each auction. To this end, this paper analyzes the strategic behavior of bidders and determines the equilibrium bidding strategies for the individual auctions that constitute a series. We do this using both first- and second-price rules in an incomplete information setting where the bidders are uncertain about their budget constraints.
Shaheen Fatima
A Comparative Study of Argumentation- and Proposal-Based Negotiation
Summary
Recently, argumentation-based negotiation has been proposed as an alternative to classical mechanism design. The main advantage of argumentation-based negotiation is that it allows agents to exchange complex justification positions rather than just simple proposals. Its proponents maintain that this property of argumentation protocols can lead to faster and beneficial agreements when used for complex multiagent negotiation. In this paper, we present an empirical comparison of argumentation-based negotiation to proposal-based negotiation in a strategic two-player scenario, using a game-theoretic solution as a benchmark, which requires full knowledge of the stage games. Our experiments show that in fact the argumentation-based approach outperforms the proposal-based approach with respect to the quality of the agreements found and the overall time to agreement.
Angelika Först, Achim Rettinger, Matthias Nickles
The Blind Leading the Blind: A Third-Party Model for Bilateral Multi-issue Negotiations under Incomplete Information
Summary
We study a multi-issue negotiation problem where agents have private information concerning their preferences over the issues. The ignorance of agents regarding the actual solution space makes it difficult for them to come to an agreement that is both fair and efficient. To make such negotiations easier, we propose a framework that employs a third-party to act as a mediator that will guide agents towards equitable solutions on the efficient frontier. To achieve this, our mediator combines the declarations of agents into a coherent negotiation protocol that dampens the desire of agents to lie and encourages them to explore regions of the solution space that are efficient and profitable for both parties.
James Shew, Kate Larson
Using Clustering Techniques to Improve Fuzzy Constraint Based Automated Purchase Negotiations
Summary
Fuzzy constraint based approaches to automated negotiation provide a negotiation framework that has been applied in automated purchase negotiation scenarios. One of the key issues that these negotiation scenarios may have to address is the inclusion of catalogue of products in the negotiation model. To this end, this chapter presents a fuzzy constraint based negotiation framework, applicable in electronic market scenarios, where seller agents own private catalogues of products, and buyer agents model their preferences by means of fuzzy constraints. In the negotiation model proposed, interactions among agents are formalized as a dialogue game protocol, where the key mechanism is the use of detailed relaxation requests. The objective of a relaxation request is to conduct the negotiation dialogue to an optimal search space. However, the generation of relaxation requirements is difficult to manage, and involves several input parameters that must be considered. A novel mechanism is proposed in order to generate relaxation requests, that is based on the use of clustering applied over the catalogue of products. We show how the performance of the negotiation processes in terms of computation time and joint utility can be improved. Specifically, via empirical evaluation, the negotiation algorithm can lead to a 35% improvement in the duration of the negotiation dialogues, and to a significant improvement in the utility of the deals that are made.
Miguel A. Lopez-Carmona, Ivan Marsa-Maestre, Juan R. Velasco, Enrique de la Hoz
Assess Your Opponent: A Bayesian Process for Preference Observation in Multi-attribute Negotiations
Summary
In an agent based multi-attribute negotiation, preferences are private knowledge. Once public, an agent’s preferences will be exploited by opposing parties trying to improve their own utility gains. Therefore, an agent will strictly hide its attribute weights from any competitor. The paper presents a method of learning an opposing party’s preferences by observing its behavior in a negotiation. The method includes a Bayesian learning process where the agent forms hypotheses based on its analysis of offers it receives. As a result, the agent improves estimation of the opponent’s preferences. Our experimental study on the performance of this negotiation strategy shows that not only does it improve the chances of the negotiation being successful at all (by a higher chance of finding the “area of agreement”), but it will also improve the individual success of the agent that applies it. The Bayesian agents perform superior to agents with static knowledge. In an all-Bayesian negotiation, there is a higher probability of a contract being reached than in a negotiation with static participation or in an all-static negotiation. Limited information about the opponent is one of the key limiting factors to successful automated negotiations. The idea of gaining knowledge that is intently kept hidden by another party is therefore a crucial basis for future automated negotiation systems that are trusted to succeed.
Christoph Niemann, Florian Lang
Designing Risk-Averse Bidding Strategies in Sequential Auctions for Transportation Orders
Summary
Designing efficient bidding strategies for agents participating in multiple, sequential auctions remains an important challenge for researchers in agent-mediated electronic markets. The problem is particularly hard if the bidding agents have complementary (i.e. super-additive) utilities for the items being auctioned, such as is often the case in distributed transportation logistics. This paper studies the effect that a bidding agent’s attitude towards taking risks plays in her optimal, decision-theoretic bidding strategy. We model the sequential bidding decision process as an MDP and we analyze, for a category of expectations of future price distributions, the effect that a bidder’s risk aversion profile has on her decision-theoretic optimal bidding policy. Next, we simulate the above strategies, and we study the effect that an agent’s risk aversion has on the chances of winning the desired items, as well as on the market efficiency and expected seller revenue. The paper extends the results presented in our previous work (reported in [1]), not only by providing additional details regarding the analytical part, but also by considering a more complex and realistic market setting for the simulations. This simulation setting is based on a real transportation logistics scenario [2]), in which bidders have to choose between several combinations (bundles) of orders that can be contracted for transportation.
Valentin Robu, Han La Poutré
CPN-Based State Analysis and Prediction for Multi-agent Scheduling and Planning
Summary
In Agent Based Scheduling and Planning Systems, autonomous agents are used to represent enterprises and operating scheduling/planning tasks. As application domains become more and more complex, agents are required to handle a number of changing and uncertain factors. This requirement makes it necessary to embed state prediction mechanisms in Agent Based Scheduling and Planning Systems. In this chapter, we introduce a Colored Petri Net based approach that use Colored Petri Net models to represent relative dynamic factors of scheduling/planning. Furthermore, in our approach, we first introduce and adopt an improved Colored Petri Net model which can not only analyse future states of a system but also estimate the success possibility of reaching a particular future state. By using the improved Colored Petri Net model, agents can predict the possible future states of a system and risks of reaching those states. Through embedding such mechanisms, agents can make more rational and accurate decisions in complex scheduling and planning problems.
Quan Bai, Fenghui Ren, Minjie Zhang, John Fulcher
Adaptive Commitment Management Strategy Profiles for Concurrent Negotiations
Summary
Since computationally intensive applications may often require more resources than a single computing machine can provide in one administrative domain, bolstering resource co-allocation is essential for realizing the Grid vision. Given that resource providers and consumers may have different requirements and performance goals, successfully obtaining commitments through concurrent negotiations with multiple resource providers to simultaneously access several resources is a very challenging task for consumers. The contribution of this work is devising a concurrent negotiation mechanism that (i) coordinates multiple one-to-many concurrent negotiations between a consumer and multiple resource providers, and (ii) manages (de-)commitments (intermediate) contracts between consumers and providers. Even though the mechanism in this work allows agents to decommit intermediate contracts by paying a penalty, it is shown that the decommitment mechanism is non-manipulable. In this paper, (i) three classes of commitment strategies for concurrent negotiation and (ii) a fuzzy decision making approach for deriving adaptive commitment management strategy profiles of a consumer are presented. Two series of experiments were carried out in a variety of settings. The first set of empirical results provide guidelines for adopting the appropriate class of commitment strategies for a given resource market. In the second set of experiments, consumer agents negotiated in n markets to acquire n resources where the market type for each resource is unknown to consumers (market types are defined by different supply and demand patterns of resources). Favorable results in the second set of experiments show that commitment management strategy profiles for a consumer derived using the fuzzy decision making approach achieved the highest expected utilities among all classes of commitment management strategy profiles.
Kwang Mong Sim, Benyun Shi
Analyses of Task Allocation Based on Credit Constraints
Summary
This paper presents a new contract model of trading with outsourcer agents and developer agents in large-scale software system manufacture. We consider a situation where ordering party does not order making software directly. Large-scale software consists of some modules. If the scale of a module is biggish, the software can be efficiently developed ordering as divided modules to some software developers. Generally, software developer has some risks as a company, such as, bankruptcy and dishonor. In such situation, it is important for an outsourcer to know how to reduce a rate of risks. In this paper, we propose a new risk diversification method of contracts with software developers in dividable software systems. In our protocol, we employ a payment policy of initial payment of and incentive fee. Then, the payment amount of initial fee is based on the developer’s credit. Thus, our protocol prevents an outsourcer from risk on the project. Further we propose a distributed task model to reduce time of development. The results of experiments show the effective strategy for ordering party where risk and number of developers change. Our simulation shows that the outsourcer can get much earnings and performance selling/using the software at an early date when the number of modules and developers increase.
Yoshihito Saito, Tokuro Matsuo
Erratum: The Prediction of Partners’ Behaviors in Self-interested Agents
No Abstract
Correct title: Effects of Revealed Area Based Selection Method for Representative-Based Protocol
Correct author information:
Katsuhide Fujita1 , Takayuki Ito2 and Mark Klein3
1 School of Techno-Business Administration, Nagoya Institute of Technology, Gokiso, Showa-ku, Nagoya 466-8555, Japan, fujita@itolab.mta. nitech.ac.jp
2 School of Techno-Business Administration/Dept. of Computer Science, Nagoya Institute of Technology, Gokiso, Showa-ku, Nagoya 466-8555, Japan / Sloan School of Management, Massachusetts Institute of Technology, Three Cambridge Center, NE20-336 Cambridge, MA 02142, USA, ito.takayuki@nitech.ac.jp
3 Sloan School of Management, Massachusetts Institute of Technology, Three Cambridge Center, NE20-336 Cambridge, MA 02142, USA, mklein@mit.edu
Fenghui Ren, Minjie Zhang
Backmatter
Metadaten
Titel
Advances in Agent-Based Complex Automated Negotiations
herausgegeben von
Takayuki Ito
Minjie Zhang
Valentin Robu
Shaheen Fatima
Tokuro Matsuo
Copyright-Jahr
2009
Verlag
Springer Berlin Heidelberg
Electronic ISBN
978-3-642-03190-8
Print ISBN
978-3-642-03189-2
DOI
https://doi.org/10.1007/978-3-642-03190-8

Premium Partner

    Bildnachweise