1、ArtificialIntelligenceCommunications 1202Group 7ArtificialIntelligenceSearch enginesLaborAppliancesSciencecontent Definition Meanings History and greatmen Developing and difficulties Knowledge Current situation Applications and Achievements Future Definition定义Artificial Intelligence (AI) is the inte
2、lligence of machines and the branch of computer science that aims to create it.Definition in AI textbook :”the study and design of intelligent agents”人工智能(AI)是机器智能与计算机科学,旨在创建它的分支。在AI教科书的定义:“学习的智能代理的设计。” 人工智能是计算机科学的前沿,充满机遇和挑战。 “A student in physics might reasonably feel that all the good ideas have a
3、lready been taken by Galileo,Newton,Einstein,and the rest,and that it takes many years of study before one can contribute new ideas,AI,on the other hand,still has openings for a full-time Einstein.” _Artificial Intelligence:A modern Approach (一位在物理学领域的学生会理所当然的认为所有的好点子已经被伽利略,牛顿,爱因斯坦和其他人,它需要许多年前的研究能做出
4、贡献的新思路,另一方面,AI,仍然作为一个全职的爱因斯坦。)“Science of making machines do things that would require intelligence if done by man.”让机器做的事情,需要智慧,如果做了人的让机器做的事情,需要智慧,如果做了人的科学。科学。 Minsky“make computer more useful and to understand principles that make intelligence possible.” 使电脑更加有用和明白原则,使智力成为可能。 WinstonThe exciting n
5、ew effort to make computers thinks machine with minds, in the full and literal sense” (Haugeland 1985)“The art of creating machines that perform functions that require intelligence when performed by people” (Kurzweil, 1990)“The study of mental faculties through the use of computational models” (Char
6、niak et al. 1985)A field of study that seeks to explain and emulate intelligent behavior in terms of computational processes” (Schalkol, 1990)Systems that think like humansSystems that think rationallySystems that act like humansSystems that act rationallyViews of AI fall into four categories:What i
7、s AI?国际人工智能联合会议国际人工智能联合会议The History of AI Secondary Translation - Machine Translation the spirit is willing but the flesh is weakthe vodka is good but the meat is rottenCombinatorial explosion problem (组合爆炸组合爆炸)The fact that a program can find a solution in principle does not means that the program
8、 contains any of the mechanisms needed to find it in practice.Perceptron limitations(感知机局限性感知机局限性) A two-input perceptron cannot be trained to recognize when its two inputs are different.Difficulties The American scholar McCarthy who is considered the father of artificial intelligence of organized a
9、 Society, many experts and scholars gathered together for a discussion of up to two months. Inviting them to participate in the Summer Dartmouth Artificial Intelligence Research Society. Since then, this field is named artificial intelligence In machine learning, in 1957 Rosenblatt (罗森布拉特 )successfu
10、lly developed Perceptron(感知机);在1957年机器产生时代,罗森布拉特成功的研制出感知机。achievement In theorem proving, in 1958 Wang Hao in the IBM-704 machine with a 3-5 minute proved all 220 theorem “Principia Mathematica” in the relevant propositional calculus;In 1965, Robinson made due to the resolution principle(归结原理);定理证明,
11、1958年王皓用IBM-704机用3-5分钟的证明所有220定理“数学原理”中的相关命题演算;1965年,罗宾逊作出了应有的决议原则。 In pattern recognition(模式识别), 1959 Selfridge(塞尔福里奇)launcheda pattern recognition program; 1965 Roberts prepared a distinguished building block construction procedures(可分辨积木构造程序);1959年塞尔弗里奇用模式识别推出了模式识别程序; 1965年罗伯茨准备了杰出的建筑模块结构的程序。In t
12、erms of problem solving(问题求解): 1960 Newell(纽厄尔) compiled a universal problem solver (GPS), can solve 11 different types of problems; In expert systems(专家系统): the results of the 1968 Feigenbaum(费根鲍姆) developed DENDRAL expert system and put into use; achievement 在专家系统:在专家系统:1968的结果费根鲍姆开发的结果费根鲍姆开发DENDR
13、AL“专家系专家系统统”并投入使用;并投入使用;在解决问题方面:在解决问题方面:1960纽厄尔编译一个普遍的问题解决者纽厄尔编译一个普遍的问题解决者(GPS),可以解决可以解决11个不同类型的问题个不同类型的问题;May 1997, the U.S. IBMs “Deep Blue” supercomputer beat the world chess champion Garry Kasparov(卡斯帕罗夫). The great achievements of artificial intelligence1997五月,美国IBM的“深蓝”超级计算机击败了国际象棋世界冠军卡斯帕罗夫(卡斯
14、帕罗夫)。Computational Intelligence thriving (90s-now)Artificial Neural Network人工神经网络人工神经网络Genetic Algorithm遗传遗传算法算法Fuzzy Reasoning模糊推理模糊推理Particle Swarm Optimization粒子群优化粒子群优化Artificial Immune System人工免疫系统人工免疫系统Artificial intelligence (人工智能)Turing test (图灵测试)pattern recognition(模式识别)expert systems(专家系统
15、)Neural Network(神经网络)The Turing test (1936)nAlan turingAlan turing(1912-1954)nThe Turing test is a test of a machines ability to exhibit intelligent behavior or indistinguishable from that of a human. If the judger cannot reliably tell the machine from the human, the machine is said to have passed t
16、he test.n图灵测试是一种测试机器的能力,表现出人类的智能行为或难以区分的。如果测试的人不能明确地区分出人类和机器,这台机器就通过了测试。Applications The main areas General machine intelligence, conversationl behavior ,data-mining, 一般机器智能一般机器智能,对话行为对话行为,数据挖掘数据挖掘 Driverless cars, robot soccer and games 无人驾驶汽车无人驾驶汽车,机器人足球和游戏机器人足球和游戏RobotsoccerConversationalbehavior
17、A comprehensive survey paper provides a thorough review of intelligent systems in process engineering. The process control research in this area has largely been concerned with three AI methods:knowledge-based systems,neural nerworks and fuzzy logic.一个广泛的调查提供了一个全面的关于智能技术在工程进程中的一系列问题的复查,针对此领域的这项进程控制调
18、查研究主要集中在三个方面:知识库系统、神经网络和模糊逻辑。AImethods:knowledge-basedsystems,NeuralnetworksandFuzzylogicprovides人工智能主要集中在三个方面:知识库系统、神经网络和模糊逻辑 Knowledge-based systems(KBS),also referred to as expert systems,use a set of “rules” to perform logical inferences about the state of a process operation or some other activ
19、ityof interest.Industrial applications of KBS systems have largely been concerned with supervisory,which have included the following problems: complex control schemes ,recovery from extreme conditons and emergency shutdowns.知识库系统(KBS)也就是专家系统,就是应用一系列的规则对操作过程的状态或其他关心的行为作出逻辑推理。知识库系统在工业上主要应用在管理活动上。管理控制的
20、应用包括这样一些问题:复杂管理计划、从极端条件和紧急停工的恢复。Neural networks provide a powerful approach for developing empirical nonlinear models for variety of physical phenomena. Neural networks also provide a promising approach for pattern recognition problems such as senor data analysis and faulty detection where tradition
21、al modeling techniques are not easily applied.神经网络为发展大量物理现象的非线性经验模型提供了有效的途径。神经网络同样为模式识别问题提供了非常有前景的途径,如传感数据分析、错误检测等这些传统模型已不能轻松应用的领域。Fuzzy logic provides conceptual framework for practical problems where some process variable are represented as “linguistic variables” which have only a few possible val
22、ues (e.g. very large ,large ,normal ,small etc.). Thus applications of fuzzy logic and fuzzy control can be viewed as special cases of KBS,which have fuzzy boundaries for the rules.模糊逻辑为实际问题提供了一个概念上的框架。在这些实际问题中,过程的变化被描述为语言上的变化,这些变化仅仅只有一些可能值(如很大、大、正常、小等)。因此,模糊逻辑和模糊控制就可看作是知识库系统的特例,其规则之间的界限是模糊的。Meaning
23、ofAI The most important fields of research in this area are information processing ,pattern recognition ,game-playing computer ,and applied fields such as medical diagnosis. Present research in information processing deals with programs that enable a computer to understand written or spoken informat
24、ion and to produce summaries ,answer specific questions or redistribute information to users interested in specific areas of this information . 对人工智能的研究体现在几个最重要的领域:信息处理、模式 识 别、博 弈计算 机和 医疗诊断。当 前对信息处理的 研究就是通过编程使计算机能够识别手写和语音信息、对信息做出概括、回答特定问题,并将信息重分配给对该信息的特定部分感兴趣的用户。 In medicine ,programs have been deve
25、loped that analyze the disease symptoms ,medical history ,and laboratory test results of a patient ,and then suggest a diagnosis to the physician .The diagnostic program is an example of so-called expert systemsprograms designed to perform takes in specialized areas as a human would .在医学上,程序还发展到根据病人的的症状、药物史和实验室检查的结果,给医师提出诊断建议。这个诊断程序就是所谓的专家系统的一个实例,专家系统就是在特殊领域可以像人一样完成工作的程序。