Evolving Fuzzy Rules for Relaxed-Criteria Negotiation

In the literature on automated negotiation, very few negotiation agents are designed with the flexibility to slightly relax their negotiation criteria to reach a consensus more rapidly and with more certainty. Furthermore, these relaxed-criteria negotiation agents were not equipped with the ability to enhance their performance by learning and evolving their relaxed-criteria negotiation rules. The impetus of this work is designing market-driven negotiation agents (MDAs) that not only have the flexibility of relaxing bargaining criteria using fuzzy rules, but can also evolve their structures by learning new relaxed-criteria fuzzy rules to improve their negotiation outcomes as they participate in negotiations in more e-markets. To this end, an evolutionary algorithm for adapting and evolving relaxed-criteria fuzzy rules was developed. Implementing the idea in a testbed, two kinds of experiments for evaluating and comparing EvEMDAs (MDAs with relaxed-criteria rules that are evolved using the evolutionary algorithm) and EMDAs (MDAs with relaxed-criteria rules that are manually constructed) were carried out through stochastic simulations. Empirical results show that: 1) EvEMDAs generally outperformed EMDAs in different types of e-markets and 2) the negotiation outcomes of EvEMDAs generally improved as they negotiated in more e-markets.      

关联规则挖掘综述

关联规则挖掘则是数据挖掘中最重要的分支之一。它着重研究大量数据中项集之间有趣的关联或相关关系，一个典型的例子就是购物篮分析。该过程可以分析出哪些商品顾客倾向于在一起购买，从而可以为商店经理提供比较好的商店布局方式。例如，通过分析，我们发现，顾客在购买了一台计算机以后，一般都会去购买财务管理软件，那么我们就可以把计算机和财务管理软件放在比较近的位置，以增加销售量。这里主要介绍了关联规则挖掘的经典算法，Apriori算法，同时给出了关联规则中的基本概念，然后分析了算法的运行效率。提出了改进的方法。     

关联规则挖掘综述

关联规则挖掘则是数据挖掘中最重要的分支之一。它着重研究大量数据中项集之间有趣的关联或相关关系,一个典型的例子就是购物篮分析。该过程可以分析出哪些商品顾客倾向于在一起购买,从而可以为商店经理提供比较好的商店布局方式。例如,通过分析,我们发现,顾客在购买了一台计算机以后,一般都会去购买财务管理软件,那么我们就可以把计算机和财务管理软件放在比较近的位置,以增加销售量。这里主要介绍了关联规则挖掘的经典算法,Apriori算法,同时给出了关联规则中的基本概念,然后分析了算法的运行效率,提出了改进的方法。     

Dynamics of rules internalized in dynamic cognitive agents playing a multi-game

Rules such as laws, institutions, and norms can be changed dynamically in our society, because they are shaped by interactions among social members who are affected by them. However, there are also some stable rules enhanced by interactions among rules. In this article, we discuss whether or not rules can be stabilized by interactions among the rules. To investigate this, we propose a multi-game model in which different games are played simultaneously by the dynamic cognitive agents. A minority game (MG) and an n-person iterated prisoners’ dilemma game (NIPDG) are adopted. In our simulation, we found that the agents internalize the complex rules expressed as intricate geometrical shapes like strange attractors on the phase spaces, when the complex macro dynamics emerged. Furthermore, it showed that the macro dynamics shaped by the macro rules in the MG can be stabilized by interaction between the MG and the NIPDG rules internalized in the agents.     

基于Adaboost算法的人脸检测

该文提出了一种基于改进的Adaboost算法的人脸检测方法.Adaboost是一种构建准确分类器的学习算法,它将一族弱学习算法通过一定规则结合成为一个强学习算法,从而通过样本训练得到一个识别准确率理想的分类器.但是,Adaboost在有高噪音样本的情况下,有可能发生过配现象,该文在Adaboost算法的基础上,对其权值更新规则做了改进,并结合PCA进行人脸检测.仿真试验表明,该方法具有良好的性能,同时可以在一定程度上有效防止过配现象的发生.     

基于IAE的缓冲滤波器参数整定与仿真

提出了一种当数字PID控制系统出现饱和时，加入缓冲滤波器改善系统动态品质的方法。基于IAE评价准则，求取缓冲滤波器参数的最优值。仿真研究表明，对目标值变化大的控制系统加入缓冲滤波器对改善因饱和现象影响系统控制品质的效果是十分明显的。     

SYMBAD—Similarity based agents for design

This paper presents the SYMBAD (similarity based agents for design) system, exploring multi-agent aspects in an architecture company, capturing, cataloging, and communicating information produced by the team members. The main task managed by the designers is to build stands to present the image of a company, project its presence in the market and emphasize the corporate identity to all prospects. From conceptual design to the construction of a final product, a stand project passes through many hands, each one adding bits and pieces until it is completed. Reuse of materials and ideas is less feasible as design complexity increases. The processes and problems in stand projects are quite common and can be easily found in other design situations. We present an agent framework to improve process awareness in an architecture company. The agents instrument the process to produce global awareness, to facilitate reuse and optimize the process as a whole. In this paper we present the agent architecture, as well as each agent’s general functioning and reasoning rules.     

SYMBAD—Similarity based agents for design

This paper presents the SYMBAD (similarity based agents for design) system, exploring multi-agent aspects in an architecture company, capturing, cataloging, and communicating information produced by the team members. The main task managed by the designers is to build stands to present the image of a company, project its presence in the market and emphasize the corporate identity to all prospects. From conceptual design to the construction of a final product, a stand project passes through many hands, each one adding bits and pieces until it is completed. Reuse of materials and ideas is less feasible as design complexity increases. The processes and problems in stand projects are quite common and can be easily found in other design situations. We present an agent framework to improve process awareness in an architecture company. The agents instrument the process to produce global awareness, to facilitate reuse and optimize the process as a whole. In this paper we present the agent architecture, as well as each agent’s general functioning and reasoning rules.     

软件测试性设计综述

软件测试性设计分为四类：设计时应当遵循的测试性设计原则,通过改变设计或代码提高软件测试性的专用测试性设计技术,为软件增加专门测试结构的结构化测试性设计技术和在软件开发全周期考虑软件测试的测试性设计综合技术。软件测试性设计方法借用了很多硬件测试性设计思想,但在许多方面还能进一步发展。     

Rights and Argumentation in Open Multi-Agent Systems

As utility calculus cannot account for an important part of agents' behaviour in Multi-Agent Systems, researchers have progressively adopted a more normative approach. Unfortunately, social laws have turned out to be too restrictive in real-life domains where autonomous agents' activity cannot be completely specified in advance. It seems that a halfway concept between anarchic and off-line constrained interaction is needed. We think that the concept of right suits this idea. Rights improve coordination and facilitate social action in Multi-Agent domains.Rights allow the agents enough freedom, and at the same time constrain them (prohibiting specific actions). Besides, rights can be understood as the basic concept underneath open normativesystems where the agents reason about the code they must abide by. Typically, in such systems this code is underspecified. On the other hand, the agents might not have complete knowledge about the rules governing their interaction. Conflict situations arise, thus, when the agents have different points of view as to how to apply the code. We have extended Parsons's et al. argumentation protocol (Parsons et al. 1998a, b) to normative systems to deal with this problem.     

Agents that Learn How to Generate Arguments from Other Agents

Learning how to argue is a key ability for a negotiator agent. In this paper, we propose an approach that allows agents to learn how to build arguments by observing how other agents argue in a negotiation context. Particularly, our approach enables the agent to infer the rules for argument generation that other agents apply to build their arguments. To carry out this goal, the agent stores the arguments uttered by other agents and the facts of the negotiation context where each argument is uttered. Then, an algorithm for fuzzy generalized association rules is applied to discover the desired rules. This kind of algorithm allows us (a) to obtain general rules that can be applied to different negotiation contexts; and (b) to deal with the uncertainty about the knowledge of what facts of the context are taken into account by the agents. The experimental results showed that it is possible to infer argument generation rules from a reduced number of observed arguments.     

Incorporating target language semantic roles into a string-to-tree translation model

The string-to-tree model is one of the most successful syntax-based statistical machine translation (SMT) models. It models the grammaticality of the output via target-side syntax. However, it does not use any semantic information and tends to produce translations containing semantic role confusions and error chunk sequences. In this paper, we propose two methods to use semantic roles to improve the performance of the string-to-tree translation model: (1) adding role labels in the syntax tree; (2) constructing a semantic role tree, and then incorporating the syntax information into it. We then perform string-to-tree machine translation using the newly generated trees. Our methods enable the system to train and choose better translation rules using semantic information. Our experiments showed significant improvements over the state-of-the-art string-to-tree translation system on both spoken and news corpora, and the two proposed methods surpass the phrase-based system on large-scale training data.     

Crowds by Example

We present an example-based crowd simulation technique. Most crowd simulation techniques assume that the behavior exhibited by each person in the crowd can be defined by a restricted set of rules. This assumption limits the behavioral complexity of the simulated agents. By learning from real-world examples, our autonomous agents display complex natural behaviors that are often missing in crowd simulations. Examples are created from tracked video segments of real pedestrian crowds. During a simulation, autonomous agents search for examples that closely match the situation that they are facing. Trajectories taken by real people in similar situations, are copied to the simulated agents, resulting in seemingly natural behaviors.     

数据时效性修复问题的求解算法

数据过时问题普遍存在于实际应用中,因此将数据库中的过时数据修复为最新值是提高数据质量的关键步骤.当前主要有基于规则和基于统计2类数据修复方法.基于规则的修复方法能够将领域知识直观地表达为规则的形式,但是难以表达数据中某些复杂的关联关系;基于统计的方法能够表达数据中的复杂关联关系,并修复许多通过规则难以发现和修复的错误,但是该类方法均需要学习较复杂的条件概率分布,且难以直接应用数据语义相关的领域知识.研究数据时效性的修复问题,同时,为了克服当前2类数据修复方法的缺点,提出一类新的修复规则,将规则和统计的方法结合起来修复过时数据.该规则一方面能够以传统规则的方式表达领域知识,另一方面还能够使用其特有的分布表来描述数据随时间变化的统计信息.接着,还给出了修复规则学习算法和数据时效性修复算法.真实和虚拟数据上的实验均验证了算法的有效性.     

Minority Games, Local Interactions, and Endogenous Networks

We study a local version of the Minority Game, where agents are placed on the nodes of a directed graph. Agents care about being in the minority of the group of agents they are currently linked to and employ myopic best-reply rules to choose their next-period state. We show that, in this benchmark case, the smaller the size of local networks, the larger long-run population-average payoffs. We then explore the collective behavior of the system when agents can: (i) assign weights to each link they hold and modify them over time in response to payoff signals; (ii) delete badly performing links (i.e., opponents) and replace them with randomly chosen ones. Simulations suggest that, when agents are allowed to weigh links but cannot delete/replace them, the system self-organizes into networked clusters that attain very high payoff values. These clustered configurations are not stable and can easily be disrupted, generating huge subsequent payoff drops. If, however, agents can (and are sufficiently willing to) discard badly performing connections, the system quickly converges to stable states where all agents get the highest payoff, independently of the size of the networks initially in place.JEL Classification:s C72, C73.     

Design and evaluation of visualization support to facilitate association rules modeling

Association rules mining is a popular data mining modeling tool. It discovers interesting associations or correlation relationships among a large set of data items, showing attribute values that occur frequently together in a given dataset. Despite their great potential benefit, current association rules modeling tools are far from optimal. This article studies how visualization techniques can be applied to facilitate the association rules modeling process, particularly what visualization elements should be incorporated and how they can be displayed. Original designs for visualization of rules, integration of data and rule visualizations, and visualization of rule derivation process for supporting interactive visual association rules modeling are proposed in this research. Experimental results indicated that, compared to an automatic association rules modeling process, the proposed interactive visual association rules modeling can significantly improve the effectiveness of modeling, enhance understanding of the applied algorithm, and bring users greater satisfaction with the task. The proposed integration of data and rule visualizations can significantly facilitate understanding rules compared to their nonintegrated counterpart.     

Fuzzy function approximation with ellipsoidal rules

A fuzzy rule can have the shape of an ellipsoid in the input-output state spare of a system. Then an additive fuzzy system approximates a function by covering its graph with ellipsoidal rule patches. It averages rule patches that overlap. The best fuzzy rules cover the extrema or bumps in the function. Neural or statistical clustering systems can approximate the unknown fuzzy rules from training data. Neural systems can then both tune these rules and add rules to improve the function approximation. We use a hybrid neural system that combines unsupervised and supervised learning to find and tune the rules in the form of ellipsoids. Unsupervised competitive learning finds the first-order and second-order statistics of clusters in the training data. The covariance matrix of each cluster gives an ellipsoid centered at the vector or centroid of the data cluster. The supervised neural system learns with gradient descent. It locally minimizes the mean-squared error of the fuzzy function approximation. In the hybrid system unsupervised learning initializes the gradient descent. The hybrid system tends to give a more accurate function approximation than does the lone unsupervised or supervised system. We found a closed-form model for the optimal rules when only the centroids of the ellipsoids change. We used numerical techniques to find the optimal rules in the general case.     

Specification and refinement of networks of asynchronously communicating agents using the assumption/commitment paradigm

This paper presents an assumption/commitment specification technique and a refinement calculus for networks of agents communicating asynchronously via unbounded FIFO channels in the tradition of Kahn.

We define two types of assumption/commitment specifications, namely simple and general specifications.

It is shown that semantically, any deterministic agent can be uniquely characterized by a simple specification, and any nondeterministic agent can be uniquely characterized by a general specification.

We define two sets of refinement rules, one for simple specifications and one for general specifications. The rules are Hoare-logic inspired. In particular the feedback rules employ invariants in the style of a traditional while-rule.

Both sets of rules have been proved to be sound and also (semantic) relative complete.

Conversion rules allow the two logics to be combined. This means that general specifications and the rules for general specifications have to be introduced only at the point in a system development where they are really needed.

     

Regularly irregular: how groups reconcile cross-cutting agendas and demand in healthcare

The flow of technical work in acute healthcare varies unpredictably, in patterns that occur regularly enough that they can be managed. Acute care organizations develop ways to hedge resources so that they are available if they are needed. This pragmatic approach to the distribution of work among and across groups shows how rules can be used to manage a response to irregular demands for care. However, no rule set can be complete enough to cover this setting’s variety of care demands. Expertise is also needed to tie together the loose ends of conflicts that remain where rules no longer suffice. Many informal solutions to systemic problems go unnoticed unless they are the subjects of study. Naturalistic decision making (NDM) methods such as observational study, interviews, and process tracing reveal the activities of workers in their natural settings. Results of findings from such explorations of technical work can improve understanding of large scale work processes and, ultimately, patient safety. We have explored how practitioners cope with the demands that the system presents to them. While not all succeed, successful initiatives workers have developed demonstrate how their solutions create resilience at large scale.     

Mapping multi-spectral remote sensing images using rule extraction approach

To improve the accurate rate of mapping multi-spectral remote sensing images, in this paper we construct a class of HyperRectangular Composite Neural Networks (HRCNNs), integrating the paradigms of neural networks with the rule-based approach. The supervised decision-directed learning (SDDL) algorithm is also adopted to construct a two-layer network in a sequential manner by adding hidden nodes as needed. Thus, the classification knowledge embedded in the numerical weights of trained HRCNNs can be extracted and represented in the form of If-Then rules. The rules facilitate justification on the responses to increase accuracy of the classification. A sample of remote sensing image containing forest land, river, dam area, and built-up land is used to examine the proposed approach. The accurate recognition rate reaching over 99% demonstrates that the proposed approach is capable of dealing with image mapping.