科学创造力的机器发现研究述评

机器发现是结合科学哲学、科学史、人工智能和认知心理学对科学创造力进行研究的领域。这个跨学科的研究进路有4个主要研究方向。作为创造力研究的新进路,机器发现研究从方法、视角和对象等方面推进了对科学创造思维的探索,也因其方法论和认识论的特征引发了争议与挑战。本文在对上述诸方面进行分析的基础上,也对该研究进路的未来发展进行了思考。     

论古代科学发现优先权的确立原则与判定标准——以勾股定理与毕达哥拉斯定理的发现为例

通过对勾股定理与毕达哥拉斯定理发现与证明方法比较,考证了中西方对勾股定理的发现均符合现代科学发现的定义,提出了中国与西方几乎同时独立发现勾股定理。根据时间敏感性差异原理,指出了古代科学发现优先权的确立原则和判定标准,中西方这种独立的发现开辟了中西方科学发展的不同模式。     

科学发现的本质及其逻辑机制的再发现

不管对科学发现进行语义分析，还是对科学发现活动本身进行历史探究，我们都会发现，科学发现除了具有社会、历史和心理上的属性外，它在本质上乃是新假说的产生、选择、修正、接受和解释的逻辑过程，其逻辑机制正是皮尔士与汉森等所倡导的溯因推理。而且，现代认知科学的新成果促进了溯因推理形式的不断完善，人们意识到在新假说产生与选择中背景理论与背景知识的重要作用，创新了溯因推理的推理形式，令人满意地回答了溯因推理何以能够以及如何能够成为科学发现的逻辑的问题。     

论科学发现作为问题解决的特殊性——对赫伯特·西蒙机器发现研究的分析

将科学发现视为问题解决的观点,是科学哲学与认知研究相结合的产物,促成机器发现研究方法的产生。西蒙不仅指出科学发现作为问题解决的普遍特征,还概括了科学发现区别于其他问题解决的个性特征。这些个性特征从创造性的角度将科学发现与常规科学的问题解决区分开来。从科学问题的角度看,科学发现和机器发现都是从未知到已知的科学问题解决过程,体现了不同层面的创造性,使得机器发现可以通过再发现的形式模拟科学发现。     

A New Mechanism for Transfer Between Conceptual Domains in Scientific Discovery and Education

Confronted with problems or situations that do not yield toknown theories and world views, scientists and students are alike. Theyare rarely able to directly build a model or a theory thereof. Rather,they must find ways to make sense of the circumstances using theircurrent knowledge and adjusting what is recognized in the process. Thisway of thinking, using past ways of perceiving the physical world tobuild new ones does not follow a logical path and cannot be described astheory revision. Likewise, in many situations it is awkward, indeedoften impossible, to resort to analogical reasoning to account for it.This paper presents a new mechanism, called `tunnel effect', that mayexplain, in part, how scientists and students reason while constructinga new conceptual domain. `Tunnel effect' is also contrasted withanalogical reasoning.     

Scientific Discovery: Between Incommensurability of Paradigms and Historical Continuity

Discoveries in physics imply two elements. The firstone is the belief that formal tools, already foundedin the framework of existing mathematical theories,may offer the solution to a puzzling anomaly. Thesecond one is the ability to assign a physical meaningto the adopted formalism, and to consider all itstheoretical implications.Discussing an historical case where the adoption of aparticular formalism represents the real motor of thecreative intuition, we mean to delineate scientificdiscovery both as a discontinuous change with respectto previous achievements and as a linear process ofknowledge enrichment.On March 1948, during the Pocono conference thatfollowed the one held in Shelter Island, Feynmananalysed the electron-photon interaction formulatingit in terms of the Lagrangian formalism. Thedevelopment of Feynman's idea draws attention to thepoint that novel theoretical discoveries may be theresult of applying existing formal tools. They may bethe result of giving different interpretations toprevious scientific thinking (according to thehermeneutical point that not even scientific textshave a single, absolute meaning but are given amultiplicity of possible readings by different peoplein different contexts).     

Rationality in the Discovery of Empirical Laws

In this paper I argue against the traditional viewthat in discovery processes there is no place forrational decisions. First I argue that some historicalprocesses in which an empirical law was developed,were rational. Second, I identify some of themethodological rules that we can follow in order to berational when constructing an empirical law. Finally,I argue that people who deny that scientific discoverycan be rational do not understand the nature ofmethodological rules.     

Analogical Reasoning and Modeling in the Sciences

This paper aims at integrating the work onanalogical reasoning in Cognitive Science into thelong trend of philosophical interest, in this century,in analogical reasoning as a basis for scientificmodeling. In the first part of the paper, threesimulations of analogical reasoning, proposed incognitive science, are presented: Gentner's StructureMatching Engine, Mitchel's and Hofstadter's COPYCATand the Analogical Constraint Mapping Engine, proposedby Holyoak and Thagard. The differences andcontroversial points in these simulations arehighlighted in order to make explicit theirpresuppositions concerning the nature of analogicalreasoning. In the last part, this debate in cognitivescience is applied to some traditional philosophicalaccounts of formal and material analogies as a basisfor scientific modeling, like Mary Hesse`s, and tomore recent ones, that already draw from the work inArtificial Intelligence, like that proposed byAronson, Harré and Way.     

英汉科技翻译中的术语定名规则探讨

根据参与翻译评审英汉信息技术标准及辞书的体验，本文总结出在译审规范与术语定名时应注意的12条规则——基于规范；直译首选；名实照应；力避既占；类、数一致；弃繁就简；体察词外；勿宽勿严；全面寻词；求助经典；并列术语同构；统一各术语元。每条规则都举出两三个实例，加以查证、剖析和评点。本文旨在探讨科技术语的翻译和定名规则，并希望有助于探索英汉人助机译或机助人译科技文献的定名与选词决策。     

科技史研究中的"双重标准"现象

科技史研究中存在着对同类问题的判定采用“双重标准”的现象。同一作者的“双重标准”是逻辑欠严密,首尾不一贯;不同作者之间存在的“双重标准”现象则是学术观点不同的结果。“双重标准”不利于科学发现优先权的评价,需要通过学术讨论统一认识,这至标准统一。     

Model-Based and Manipulative Abduction in Science

What I call theoretical abduction (sentential and model-based)certainly illustrates much of what is important in abductive reasoning, especially the objective of selecting and creating a set of hypotheses that are able to dispense good (preferred) explanations of data, but fails to account for many cases of explanation occurring in science or in everyday reasoning when the exploitation of the environment is crucial. The concept of manipulative abduction is devoted to capture the role of action in many interesting situations: action provides otherwise unavailable information that enables the agent to solve problems by starting and performing a suitable abductive process of generation or selection of hypotheses. Many external things, usually inert from the epistemological point of view, can be transformed into what I callepistemic mediators, which are illustrated in the last part of the paper, together with an analysis of the related notions of ``perceptual and inceptual rehearsal' and of ``external representation'.     

Visual Models in Analogical Problem Solving

Visual analogy is believed to be important in human problem solving. Yet, there are few computational models of visual analogy. In this paper, we present a preliminary computational model of visual analogy in problem solving. The model is instantiated in a computer program, called Galatea, which uses a language for representing and transferring visual information called Privlan. We describe how the computational model can account for a small slice of a cognitive-historical analysis of Maxwell’s reasoning about electromagnetism.     

Covariance/Invariance: A Cognitive Heuristic in Einstein's Relativity Theory Formation

Relativity Theory by Albert Einstein has been so far littleconsidered by cognitive scientists, notwithstanding its undisputedscientific and philosophical moment. Unfortunately, we don't have adiary or notebook as cognitively useful as Faraday's. But physicshistorians and philosophers have done a great job that is relevant bothfor the study of the scientist's reasoning and the philosophy ofscience. I will try here to highlight the fertility of a `triangulation'using cognitive psychology, history of science and philosophy of sciencein starting answering a clearly very complex question:why did Einstein discover Relativity Theory? Here we arenot much concerned with the unending question of precisely whatEinstein discovered, that still remains unanswered, for we have noconsensus over the exact nature of the theory's foundations(Norton 1993). We are mainly interested in starting to answer the`how question', and especially the following sub-question: what(presumably) were his goals and strategies in hissearch? I will base my argument on fundamental publications ofEinstein, aiming at pointing out a theory-specific heuristic, settingboth a goal and a strategy: covariance/invariance.The result has significance in theory formation in science, especiallyin concept and model building. It also raises other questions that gobeyond the aim of this paper: why was he so confident in suchheuristic? Why didn't many other scientists use it? Where did he keep ? such a heuristic? Do we have any other examples ofsimilar heuristic search in other scientific problemsolving?     

中国20世纪30年代科学化运动与现实启迪

20世纪30年代科学化运动是科学家们自觉地实践科教兴国的一次尝试.他们不但从事自己的科学研究事业,还做民众的科普工作,把科学与国家工业化结合起来.他们注重科学技术的应用,但反对只把科学技术当成满足某种实际需要的工具,他们将以培养科学精神作为科学教育的使命.这个时代的科学家直接受现代科学的洗礼,但不失知识分子的人文关怀,具有传统的人文修养和气质.     

高校教师科研激励机制的构建

在高校科研管理中,一个重要的方面是建立科学合理的激励机制,调动科研人员的积极性,以充分发挥高校在国家科研事业方面的重要作用。本文对借用卓越管理模式构建高校科研激励机制进行了探讨。     

中国古代多臂织机再研究

中国古代多臂织机可加挂的综蹑数量，特别是《西京杂记》一书所载120蹑织机是否曾出现过，纺织史界的看法殊不一致。解析该类型织机的开口机构、机上经纱的拉伸变形，综蹑数量达到120时，织机仍能正常开口，《西京杂记》所载，来源有据，非不实之谈，故历史上应出现过120蹑织机。这种织机需二人联手操作，工效甚低，是多劈织机发展到极限的一种特殊机型，对提花机的发展有一定的积极影响。     

科学名词审查会

科学名词审查会是民国时期一个准官方科学名词审查组织,其前身是医学名词审查会。本文详细梳理了医学名词审查会和科学名词审查会的历次正式名词审查大会。     

数学中的经验:认识论上的紧张视角

就形式数学来说,它是脱离经验,在内涵上是一种社会建构。计算机技术提高了人们的计算能力,对数学的这一社会建构平添了紧张,加强了“作为计算的数学”与“作为证明的数学”之间的不平衡性。在历史上,相似的认识论上的争论表现在计算或实践的印度数学与精神或形式的西方数学之间。我们认为,认识论上的紧张,可通过以下方式得以消解:认同数学是基于经验、可错的观点,并据此引领数学课程标准的基本理念。