Recent Advances in Predictive (Machine) Learning

Prediction involves estimating the unknown value of an attribute of a system under study given the values of other measured attributes. In prediction (machine) learning the prediction rule is derived from data consisting of previously solved cases. Most methods for predictive learning were originated many years ago at the dawn of the computer age. Recently two new techniques have emerged that have revitalized the field. These are support vector machines and boosted decision trees. This paper provides an introduction to these two new methods tracing their respective ancestral roots to standard kernel methods and ordinary decision trees.     

风险决策背景下的预警原则争议及启示

风险预警原则能否作为风险评估中的决策规则一直存在争议。尽管表述不确定性和决策专断性成为风险预警原则主要缺陷,但在风险不确定情形下,以科学为基础的评估方法却导致理性悖论。风险评估的社会建构性,揭示出风险预防是一个多元、开放、协商的过程。风险预警原则的政策含义在于,将风险不确定性这一难题转换为可以协商的话题,以形成共识。     

Contemporary Epistemic Logic and the Lockean Thesis

This paper studies the Lockean thesis from the perspective of contemporary epistemic logic. The Lockean thesis states that belief can be defined as ‘sufficiently high degree of belief’. Its main problem is that it gives rise to a notion of belief which is not closed under conjunction. This problem is typical for classical epistemic logic: it is single-agent and static. I argue that from the perspective of contemporary epistemic logic, the Lockean thesis fares much better. I briefly mention that it can successfully be extended from single-agent to multi-agent settings. More importantly, I show that accepting the Lockean thesis (and a more sophisticated version for conditional beliefs) leads to a significant and unexpected unification in the dynamic behavior of (conditional) belief and high (conditional) probability with respect to public announcements. This constitutes a methodological argument in favor of the Lockean thesis. Furthermore, if one accepts Baltag’s Erlangen program for epistemology, this technical observation has even stronger philosophical implications: because belief and high probability display the same dynamic behavior, it is plausible that they are indeed one and the same epistemic notion.     

Inconsistencies in Constituent Theories of World Views: Quantum Mechanical Examples

We put forward the hypothesis that there exist three basic attitudes towards inconsistencies within world views: (1) The inconsistency is tolerated temporarily and is viewed as an expression of a temporary lack of knowledge due to an incomplete or wrong theory. The resolution of the inconsistency is believed to be inherent to the improvement of the theory. This improvement ultimately resolves the contradiction and therefore we call this attitude the ‘regularising’ attitude; (2) The inconsistency is tolerated and both contradicting elements in the theory are retained. This attitude integrates the inconsistency and leads to a paraconsistent calculus; therefore we will call it the paraconsistent attitude. (3) In the third attitude, both elements of inconsistency are considered to be false and the ‘real situation’ is considered something different that can not be described by the theory constructively. This indicates the incompleteness of the theory, and leads us to a paracomplete calculus; therefore we call it the paracomplete attitude. We illustrate these three attitudes by means of two ‘paradoxical’ situations in quantum mechanics, the wave-particle duality and the situation of non locality. This revised version was published online in July 2006 with corrections to the Cover Date.     

The General Concept of Antinomicity

Antinomicity is not necessarily dependent on negation; there is a more general conception of antinomicity based on the fundamental idea of opposition. To study this fact is indispensable to show first that truth and falsity are independent of assertion and negation. Then it can be seen that antinomies can be found everywhere, and that some single categories are in intrinsic opposition with themselves while others are opposed to one another in pairs. An antinomic ‘manifesto’ concludes the work. This revised version was published online in July 2006 with corrections to the Cover Date.     

新版《中国人民解放军军语》应突出时代特点和我军特色 ——以选词定词为例

新版《中国人民解放军军语》作为鲜明反映时代特点和我军改革建设创新发展的重要军事术语集,必须跟上时代发展、跟上斗争实践、跟上改革步伐、跟上法治进程,把习近平强军思想内容作为首要选词对象,把备战打仗的相关用语作为主要选词对象,把深化国防和军队改革成果作为基本选词对象,把新的军事法规制度表述作为常用选词对象。在选词定词过程中,应遵循时代性、系统性、专业性、实用性、稳定性等原则,注意正确选取资料、科学遴选词目、认真归纳整理,以构建完善具有时代特点和我军特色的军语体系。     

简单决策与复杂决策的比较分析及其启发意义

基于复杂性维度,本文把决策系统划分为简单决策和复杂决策两种系统;比较了两种系统中的决策在思维模式、理论背景、决策概念、研究范式、研究方法论、决策方法,以及理论适应的范围等方面的相互区别;通过比较分析,综合出"复杂决策与简单决策两种系统的本质差异","新研究范式"、"方法论"和"决策概念本身的演化"三个方面对理解"复杂决策"所具有的启发意义.     

Some New Results on Orthogonally Constrained Candecomp

The use of Candecomp to fit scalar products in the context of Indscal is based on the assumption that, due to the symmetry of the data matrices involved, two components matrices will become equal when Candecomp converges. Bennani Dosse and Ten Berge (2008) have shown that, in the single component case, the assumption can only be violated at saddle points in the case of Gramian matrices. This paper again considers Candecomp applied to symmetric matrices, but with an orthonormality constraint on the components. This constrained version of Candecomp, when applied to symmetric matrices, has long been known under the acronym Indort. When the data matrices are positive definite, or have become positive semidefinite due to double centering, and the saliences are nonnegative – by chance or by constraint –, the component matrices resulting from Indort are shown to be equal. Because Indort is also free from so-called degeneracy problems, it is a highly attractive alternative to Candecomp in the present context. We also consider a well-known successive approach to the orthogonally constrained Indscal problem and we compare, from simulated and real data sets, its results with those given by the simultaneous (Indort) approach.     

The Precautionary Principle: Scientific Uncertainty and Type I and Type II Errors

We provide examples of the extent and nature of environmental and human health problems and show why in the United States prevailing scientific and legal burden of proof requirements usually cannot be met because of the pervasiveness of scientific uncertainty. We also provide examples of how may assumptions, judgments, evaluations, and inferences in scientific methods are value-laden and that when this is not recognized results of studies will appear to be more factual and value-neutral than warranted. Further, we show that there is a "tension" between the use of the 95 percent confidence rule as a normative basis to reduce speculation in scientific knowledge and other public policy and moral concerns embodied by the adoption of a precautionary principle. Finally, although there is no precise agreement regarding what a precautionary principle might entail, we make several recommendations regarding the placement of the burden of proof and the standard of proof that ought to be required in environmental and human health matters. This revised version was published online in July 2006 with corrections to the Cover Date.     

Intelligent Choice of the Number of Clusters in K-Means Clustering: An Experimental Study with Different Cluster Spreads

The issue of determining “the right number of clusters” in K-Means has attracted considerable interest, especially in the recent years. Cluster intermix appears to be a factor most affecting the clustering results. This paper proposes an experimental setting for comparison of different approaches at data generated from Gaussian clusters with the controlled parameters of between- and within-cluster spread to model cluster intermix. The setting allows for evaluating the centroid recovery on par with conventional evaluation of the cluster recovery. The subjects of our interest are two versions of the “intelligent” K-Means method, ik-Means, that find the “right” number of clusters by extracting “anomalous patterns” from the data one-by-one. We compare them with seven other methods, including Hartigan’s rule, averaged Silhouette width and Gap statistic, under different between- and within-cluster spread-shape conditions. There are several consistent patterns in the results of our experiments, such as that the right K is reproduced best by Hartigan’s rule – but not clusters or their centroids. This leads us to propose an adjusted version of iK-Means, which performs well in the current experiment setting.     

投票悖论是严格的逻辑悖论吗?--投票悖论逻辑结构浅析

严格意义上的逻辑悖论有三大构成要素:公认正确的背景知识、严密无误的逻辑推导、能够推出矛盾等价式.投票悖论是指在群体选择的投票决策过程中,根据潜在的公共背景知识:理性人假设、传递性规则和多数规则,投票群体最后得到自相矛盾的投票结果,而这一矛盾结果是经过严密无误的逻辑推证得出的,因此投票悖论是一种典型的逻辑悖论.     

鲁滨逊能遵守规则吗?——关于个体论与集体论之争的解决

基于维特根斯坦的遵守规则理论,以鲁滨逊为例,分别论述了个体论者麦金和集体论者布鲁尔对于一个与世隔绝的鲁滨逊是否能遵守规则的看法。对于个体论者和集体论者来说,他们都认为与世隔绝的鲁滨逊能遵守规则,虽然前提一致,但产生的结论却大相径庭。对于个体论者来说,遵守规则的前提只不过是基于个体心理活动的一种意图,而对集体论者来说遵守规则必须是可共享的,它离不开社会及周围环境的影响。     

A comparison between two distance-based discriminant principles

A distance-based classification procedure suggested by Matusita (1956) has long been available as an alternative to the usual Bayes decision rule. Unsatisfactory features of both approaches when applied to multinomial data led Goldstein and Dillon (1978) to propose a new distance-based principle for classification. We subject the Goldstein/Dillon principle to some theoretical scrutiny by deriving the population classification rules appropriate not only to multinomial data but also to multivariate normal and mixed multinomial/multinormal data. These rules demonstrate equivalence of the Goldstein/Dillon and Matusita approaches for the first two data types, and similar equivalence is conjectured (but not explicitly obtained) for the mixed data case. Implications for sample-based rules are noted.     

"Białkowski Panel" - Introduction and Acknowledgements

The two main points of this contribution are the following: (1) Applied mathematical theories might complement physical theories in an essential way; some applied mathematical theories allow us to understand phenomena we are unable to explain by resorting to physical theories alone, (2) In the case of social sciences it might be necessary to account for examined phenomena by resorting to the idea of goal-oriented activity (the causal approach typical for natural science might be unsatisfactory). Weinberg's idea of grand reductionism ignores the two above mentioned facts and hence overestimates the foundational role of physics and its methodology. This revised version was published online in July 2006 with corrections to the Cover Date.     

Automated design of multiple-class piecewise linear classifiers

A new method and a supporting theorem for designing multiple-class piecewise linear classifiers are described. The method involves the cutting of straight line segments joining pairs of opposed points (i.e., points from distinct classes) ind-dimensional space. We refer to such straight line segments aslinks. We show how nearly to minimize the number of hyperplanes required to cut all of these links, thereby yielding a near-Bayes-optimal decision surface regardless of the number of classes, and we describe the underlying theory. This method does not require parameters to be specified by users — an improvement over earlier methods. Experiments on multiple-class data obtained from ship images show that classifiers designed by this method yield approximately the same error rate as the bestk-nearest neighbor rule, while providing faster decisions.This research was supported in part by the Army Research Office under grant DAAG29-84-K-0208 and in part by the University of California MICRO Program. We thank R. W. Doucette of the U.S. Naval Weapons Center and R. D. Holben of Ford Aerospace Corporation for providing the ship images in our experiments.     

欧洲技术交流协会对德国企业术语工作状况的调研报告

德国技术交流协会是欧洲最大的技术交流协会,它是在德国范围内为人们的技术交流活动提供信息交换和经验交流的最佳平台。文章介绍时任德国术语信息文献中心主任的施米茨教授和德国技术交流协会施特劳布女士在2016年牵头对德国企业开展术语工作情况的再调查。此调研报告英文版刊登在2016年9月《技术交流杂志》的英文网络版上。算是2010年同项调研的续篇。此项调研是以德国技术交流协会的名义进行的。文章旨在介绍这篇报告,为读者了解德国企业的术语工作提供一些信息。     

CLUSCALE ("CLUstering and multidimensional SCAL[E]ing"): A Three-Way Hybrid Model Incorporating Overlapping Clustering and Multidimensional Scaling Structure

Traditional techniques of perceptual mapping hypothesize that stimuli are differentiated in a common perceptual space of quantitative attributes. This paper enhances traditional perceptual mapping techniques such as multidimensional scaling (MDS) which assume only continuously valued dimensions by presenting a model and methodology called CLUSCALE for capturing stimulus differentiation due to perceptions that are qualitative, in addition to quantitative or continuously varying perceptual attributes or dimensions. It provides models and OLS parameter estimation procedures for both a two-way and a three-way version of this general model. Since the two-way version of the model and method has already been discussed by Chaturvedi and Carroll (2000), and a stochastic variant discussed by Navarro and Lee (2003), we shall deal in this paper almost entirely with the three-way version of this model. We recommend the use of the three-way approach over the two-way approach, since the three-way approach both accounts for and takes advantage of the heterogeneity in subjects’ perceptions of stimuli to provide maximal information; i.e., it explicitly deals with individual differences among subjects.     

On a Transvariation Based Measure of Group Separability

In this paper, the potentialities of transvariation (Gini, 1959) in measuring the separation between two groups of multivariate observations are explored. With this aim, a modified version of Gini’s notion of multidimensional transvariation is proposed. According to Gini (1959), two groups G₁ and G₂ are said to transvary on the k-dimensional variable X = (X₁,...,X_h,...,X_k) if there exists at least one pair of units, belonging to different groups, such that for h = 1,...,k the sign of the difference between their X_h values is opposite to that of m_1h −m_2h, where m_1h and m_2h are the corresponding group mean values of X_h. We introduce a modification that allows us to derive a measure of group separation, which can be profitably used in discriminating between two groups. The performance of the measure is tested through simulation experiments. The results show that the proposed measure is not sensitive to distributional assumptions and highlight its robustness against outliers.     

Feedback Models of Two Classical Philosophical Positions and a Semantic Problem

The notion of feedback has been exploited with considerable success in scientific and technological fields as well as in the sciences of man and society. Its use in philosophical, cultural and educational contexts, however, is still rather meagre, even if some notable attempts can be found in the literature. This paper shows that the feedback concept can help learn and understand some classical philosophical theories. In particular, attention focuses on Fichte’s doctrine of science, usually presented in obscure terms following its inventor’s style, and on the vulgate version of Hegel’s dialectic. Also a classic problem of linguistics concerning the meaning of sentences in partially unknown languages is interpreted with the aid of feedback diagrams. Even if the analysis is only qualitative, it is believed that it may serve as a useful tool for thought for both students and researchers.     

Do We Need To Search For The Only True World View?

It is necessary to take into account that every ontology and also every scientific system draws a picture of the World according to the abstractions and presuppositions which were accepted, consciously or unconsciously, during the construction of the system. That is why Aristotle, Hegel, and the paraconsistent logics gave us different world views. On the basis of contemporary logics, including paraconsistent logics, we can better understand what the objects of the Aristotelian logic are, what are the presuppositions used in it, what are the meanings of its terms and how their meaning was changed when used in other theories. This revised version was published online in July 2006 with corrections to the Cover Date.