Ahmed Moustafa (アーメド アブデル サター エルハディ ムスフタファ)

Ahmed Abdel Satar Elhady Moustafa

写真a

所属学科・専攻等

情報工学教育類 / 知能情報分野
情報工学専攻 / 知能情報分野

職名

准教授

 

論文

  • A Case-based Reasoning Approach for Automated Facilitation in Online Discussion Systems

    Wen Gu, Ahmed Moustafa, Takayuki Ito, Minejie Zhang, Chunsheng Yang

    The 13th International Conference on Knowledge Information and Creativity Support Systems(KICSS2018) ( Artificial Intelligence Association of Thailand (AIAT) )  The 13th International Conference on Knowledge Information and Creativity Support Systems(KICSS2018) ( The 13th International Conference on Knowledge Information and Creativity Support Systems(KICSS2018) ) 206 - 210   2018年11月  [査読有り]

    研究論文(国際会議プロシーディングス)   共著

    Online discussion systems have recently attracted
    great attention as an enabling approach of realizing collective
    intelligence. During online discussions, human facilitators are
    introduced in order to help these discussions to proceed more
    efficiently and productively. However, there are a number of
    challenges such as human bias and time restriction that need
    to be solved in the human facilitator-based online discussion
    systems. As a result, automated facilitation becomes necessary
    in order to overcome these shortcomings. This paper proposes
    a novel approach for automated facilitation that utilizes case
    based reasoning (CBR) in order to imitate the human facilitator

    thinking style. The proposed approach works in issue based in-
    formation system (IBIS) discussion style where complex problems

    are designed as a conversation amongst several stockholders.
    These stockholders, in turn, bring their expertise in order to
    resolve the discussion point. Experimental results show the ability
    of the proposed approach to improve the performance of online
    discussion systems, and to guide the online discussion towards
    consensus and towards gathering wisdom efficiently.

  • The Design of Meta-Strategy that Can Obtain Higher Negotiating Efficiency

    Tang Xun, Ahmed Moustafa, Takayuki Ito

    The 13th International Conference on Knowledge Information and Creativity Support Systems(KICSS2018) ( Artificial Intelligence Association of Thailand (AIAT) )  The 13th International Conference on Knowledge Information and Creativity Support Systems(KICSS2018) ( The 13th International Conference on Knowledge Information and Creativity Support Systems(KICSS2018) ) 258 - 263   2018年11月  [査読有り]

    研究論文(国際会議プロシーディングス)   共著

    The purpose of this paper is to introduce Jupiter,
    a new environment for automated negotiation in which we can
    easily create agents that are able to use machine learning in
    order to extend their knowledge. Automated negotiation is one
    of the important solutions to coordinate amongst rational agents
    with conflicting interests. In the literature, previous efforts have
    been made in order to develop a platform that is able to
    simulate automated negotiations, i.e., Genius. In this regard,
    Genius provides an environment for automated negotiation that
    aims to solve multi-issue negotiation problems. In recent years,
    with the advancement of hardware technology, the development
    of machine learning algorithms has seen a remarkable growth.
    However, there is still few research that provides support for

    developing new negotiation environments, especially these envi-
    ronments that are able to support machine learning algorithms.

    In this paper, we propose Jupiter, a new automated negotiation
    environment in which we can easily create agents that are able
    to use machine learning in order to extend their knowledge.
    In addition, we compare Jupiter with Genius and show the
    performance gains of using Jupiter.

  • Jupiter: An Automated Negotiation Environment for Supporting Agents that Use Machine Learning

    Tomoya Fukui, Ahmed Moustafa, Takayuki Ito

    The 13th International Conference on Knowledge Information and Creativity Support Systems(KICSS2018) ( Artificial Intelligence Association of Thailand (AIAT) )  The 13th International Conference on Knowledge Information and Creativity Support Systems(KICSS2018) ( The 13th International Conference on Knowledge Information and Creativity Support Systems(KICSS2018) ) 264 - 269   2018年11月  [査読有り]

    研究論文(国際会議プロシーディングス)   共著

    Given the growing interest in automated negotiation,
    the search for effective strategies has produced a variety of
    different negotiation agents. In this regard, the Automated
    Negotiating Agents Competition(ANAC) is being held annually
    since 2010. The ANAC is an international competition that
    promotes researchers to design intelligent agents that are able
    to operate effectively in several kinds of scenarios. In this
    competition, researchers analyze the negotiating agents from
    several perspectives including utility, social welfare, distance to
    Nash solution, distance to Pareto efficiency and so on. Most of
    these analyses are based on the negotiation results. In fact, the
    efficiency of the negotiation process greatly affects the negotiation
    results. In our previous work, we introduced a metric that is able
    to evaluate the efficiency of the negotiation process. In this paper,
    we propose a novel meta-strategy that utilizes this metric in order
    to obtain higher negotiating efficiency.

  • Agent33: An Automated Negotiator with Heuristic Method for Searching Bids Around Nash Bargaining Solution

    Shan Liu, Ahmed Moustafa, Takayuki Ito

    The 21st International Conference on Principles and Practice of Multi-Agent System ( Springer )  Lecture Notes in Computer Science, vol 11224 ( Lecture Notes in Computer Science, vol 11224 ) 519 - 526   2018年10月  [査読有り]

    研究論文(国際会議プロシーディングス)   共著

    The international Automated Negotiating Agents Competition (ANAC) is being held annually since 2010 in order to bring together the researchers from the multi-agent negotiation community. In this regard, the Repeated Multilateral Negotiation League (RMNL), one of the four negotiation research challenges in ANAC 2018, requires participants to design and implement an intelligent negotiating agent, that is able to negotiate with two other opponents and that is able to learn from its previous negotiation experiences. In this context, in this paper, we design a negotiating agent that focuses on searching the space of suitable bids that provide high utilities for both sides near the Nash Bargaining Solution (NBS) using a novel heuristic method. The proposed agent has participated in the ANAC competition successfully and finished in the second place in the social welfare category.

  • A Deep Reinforcement Learning Approach for Large-Scale Service Composition

    Ahmed Moustafa, Takayuki Ito

    The 21st International Conference on Principles and Practice of Multi-Agent System ( Springer, Cham )  Lecture Notes in Computer Science, vol 11224 ( Lecture Notes in Computer Science, vol 11224 ) 296 - 311   2018年10月  [査読有り]

    研究論文(国際会議プロシーディングス)   共著

    As service-oriented environments become widespread, there exists a pressing need for service compositions to cope with the high scalability, complexity, heterogeneity and dynamicity features inherent in these environments. In this context, reinforcement learning has emerged as a powerful tool that empowers adaptive service composition in open and dynamic environments. However, most of the existing implementations of reinforcement learning algorithms for service compositions are inefficient and fail to handle large-scale service environments. Towards this end, this paper proposes a novel approach for adaptive service composition in dynamic and large-scale environments. The proposed approach employs deep reinforcement learning in order to address large-scale service environments with large number of service providers. Experimental results show the ability and efficiency of the proposed approach to provide successful service compositions in dynamic and large-scale service environments.

  • A Proposal of Automated Negotiation Simulator "Jupiter" for Negotiating Agents Using Machine Learning

    Tomoya Fukui, Ahmed Moustafa, Takayuki Ito

    The 11th International Workshop on Automated Negotiation (ACAN2018) ( Springer )    2018年  [査読有り]

    研究論文(国際会議プロシーディングス)   共著

    The purpose of this paper is to propose Jupiter, a new environment for automated negotiations in which we can easily create agents that is able to use machine learning.
    Genius is cited as a prior study of the environment where automated negotiation can be simulated.
    It provides an environment for automated negotiations that aim to solve multi-issue negotiation problems.
    In the field of automated negotiation, it is expected that agreement results are optimized by machine learning.
    However, it is difficult to use Genius to simulate automated negotiations with agents using machine learning, because the past negotiations information provided by Genius is insufficient.
    As above, we propose Jupiter as a new automated negotiation environment in which we can easily create agents that is able to use machine learning.
    In addition, we compare Jupiter with Genius and show the superiority of Jupiter.

  • On Measuring the Opposition Level Amongst Intelligent Agents in Multi-issue Closed Negotiation Scenarios

    Tatsuya Toyama, Ahmed Moustafa and Takayuki Ito

    The 12th International Conference on Knowledge Information and Creativity Support Systems(KICSS2017)   The 12th International Conference on Knowledge Information and Creativity Support Systems(KICSS2017) ( The 12th International Conference on Knowledge Information and Creativity Support Systems(KICSS2017) )   2017年  [査読有り]

    研究論文(国際会議プロシーディングス)   共著

    Automated intelligent agents have been introduced by researchers in order to address multi-issue closed negotiation scenarios. These intelligent agents are able to control the negoti- ation strategies on behalf of different participant stakeholders and thus maximizing their overall utility. In this regard, it becomes crucial to quantitatively analyze the utility information of these agents that is associated with several negotiation domains and multiple negotiation scenarios. This analysis is important in enhancing the agents negotiation strategies and their ability to maximize their overall gains. Towards this end, this paper proposes a metric of the opposition level amongst several in- telligent negotiating agents. The proposed metric quantitatively analyze the utility information of participant agents in order to measure the hostility level of these agents and hence their prospective likelihood of reaching an agreement. For the purpose of evaluating the proposed metric, a set of empirical experiments is conducted with an automated negotiating simulator. The results of these experiments show the impact of the proposed metric on the agents negotiation results and hence its vital role in building better negotiation strategies.

  • An Ordering Mechanism for Automated Negotiation Among Nonlinear Utility Agents

    Tang Xun, Ahmed Moustafa, Takayuki Ito

    The 12th International Conference on Knowledge Information and Creativity Support Systems(KICSS2017)   The 12th International Conference on Knowledge Information and Creativity Support Systems(KICSS2017) ( The 12th International Conference on Knowledge Information and Creativity Support Systems(KICSS2017) )   2017年  [査読有り]

    研究論文(国際会議プロシーディングス)   共著

    There has been a growing interest of automated negation approaches in the research field of multiagent systems. Automated negotiation becomes more important when considering nonlinear utility spaces. In nonlinear utility spaces, there exist several complex relations of interdependence amongst the negotiation issues. In this regard, the aim of every negotiating agent becomes getting a higher social welfare in a shorter time considering the nonlinear utility spaces. Towards this end, this paper proposes a mechanism that enables the intelligent agents to only negotiate about one part of the negotiation issues at a time. This part is selected as the part whose negotiation issues have the most relations of interdependence amongst them. If the negotiating agents can’t reach an agreement, they negotiate about extra parts with more issues. By using this mechanism, negotiation is only for one part of the issues at a time. Hence, the question of negotiation in nonlinear spaces will be simpler. Further, the proposed mechanism introduces a concept called automatic agreement, which helps to decide the order of issues to be negotiated.

  • Trustworthy Stigmergic Service Composition and Adaptation in Decentralized Environments

    Ahmed Moustafa, Minjie Zhang , Quan Bai

    IEEE Transactions on Service Computing ( IEEE )  Vol. 9, No. 2 ( Vol. 9, No. 2 ) 317 - 329   2016年03月  [査読有り]

    研究論文(学術雑誌)   共著

    The widespread use of web services in forming complex online applications requires service composition to cope with highly dynamic and heterogeneous environments. Traditional centralized service composition techniques are not sufficient to address the needs of applications in decentralized environments. In this paper, a stigmergic-based approach is proposed to model the decentralized service interactions and handle service composition in highly dynamic open environments. In the proposed approach, web services and resources are modeled as multiple agents. Stigmergic-based self-organization mechanisms among agents are deployed to facilitate adapting service composition. In addition, to overcome the limitations of traditional QoS-based approaches, trust measurements are deployed as a criterion for service selection. To improve the performance of the proposed stigmergic-based approach under dynamic scale-free environments, we investigate the hybridization with local search operators to consolidate adaptation, and diversity schemes are introduced to facilitate continual service adaptation. Extensive experiments show the efficiency of the proposed approach in dealing with incomplete information and dynamic factors in composing and adapting web services in open environments. The experiment results also show that the proposed approach achieves a better performance than other traditional approaches.

  • Multi-Objective Service Composition in Uncertain Environments

    Ahmed Moustafa, Minjie Zhang

    IEEE Transactions on Service Computing ( IEEE )    2015年06月  [査読有り]

    研究論文(学術雑誌)   共著

    Web services have the potential to offer the enterprises with the ability to compose internal and external business services in order to accomplish complex processes. Service composition then becomes an increasingly challenging issue when complex and critical applications are built upon services with different QoS criteria. However, most of the existing QoS-aware service composition techniques are simply based on the assumption that multiple QoS criteria, no matter whether these multiple criteria are conflicting or not, can be combined into a single criterion to be optimized, according to some utility functions. In practice, this can be very difficult as these utility functions or weights are not well-known a priori. In addition, the existing approaches are designed to work in certain environments, where the QoS parameters are well-known in advance. These approaches will render fruitless when facing uncertain and dynamic environments, e.g., cloud environments, where no prior knowledge of the QoS parameters is available. In this paper, two novel multi-objective approaches are proposed to handle QoS-aware Web service composition with conflicting objectives and various restrictions on the quality matrices. The proposed approaches use reinforcement learning in order to deal with the uncertainty characteristics inherent in open and dynamic environments. Experimental results reveal the ability of the proposed approaches to find a set of Pareto optimal solutions, which have the equivalent quality to satisfy multiple QoS

全件表示 >>