一种多级评分的认知诊断模型：P-DINA模型的开发

当前绝大多数认知诊断计量模型仅适用于0-1评分数据资料, 大大限制了认知诊断在实际中的应用, 也限制了认知诊断的进一步推广和发展。本文对具有较好发展前景的DINA模型进行拓展, 开发出适合多种评分(含0-1二级评分和多级评分)数据资料的P-DINA模型, 同时采用MCMC算法实现模型参数的估计, 并对该模型性能进行研究。结果表明：(1)本文开发的P-DINA模型无论是在无结构型属性层级关系下还是在结构型属性层级关系下, 参数估计的精度均较高, 参数估计的稳健性较强, 说明开发的P-DINA模型基本合理、可行。(2)P-DINA模型可采用MCMC算法实现参数估计, 且参数估计的精度较高。(3)整体来看, 无结构型属性层级关系和结构型属性层级关系下, P-DINA模型在项目参数的估计精度上两者基本相当; 但在被试属性判准率(MMR和PMR)上无结构型属性层级关系表现的稍差一些。(4)无结构型属性阶层关系下：模型诊断的属性个数越多, 参数 估计的精度越差、属性诊断的正确率(MMR和PMR)越低, 但参数 的估计精度越好; 若想保证属性模式判准率在80%以上, 建议诊断的属性个数不宜超过7个。总之, 本研究为拓展认知诊断在教育学和心理学中的应用提供了一种新方法、新模型。     

Does Wittgenstein Actually Undermine the Foundation of Artificial Intelligence?

Wittgenstein is widely viewed as a potential critic of a key philosophical assumption of the Strong Artificial Intelligence (AI) thesis,namely,that it is in principle possible to build a programmed machine which can achieve real intelligence.Stuart Shanker has provided the most systematic reconstruction of the Wittgensteinian argument against AI,building on Wittgenstein's own statements,the "rule-following" feature of language-games,and the putative alliance between AI and psychologism.This article will attempt to refute this reconstruction and its constituent arguments,thereby paving the way for a new and amicable rather than agonistic conception of the Wittgensteinian position on AI.     

Association of the DAT1 genotype with inattentive behavior is mediated by reading ability in a general population sample

Attention deficit hyperactivity disorder (ADHD) and reading disability (RD) frequently co-occur in the child population and therefore raise the possibility of shared genetic etiology. We used a quantitative trait loci (QTL) approach to assess the involvement of the dopamine transporter (DAT1) gene polymorphism in mediating reading disability and poor attention in a general population sample of primary school children aged 6-11 years in the UK. The potential confounding effects of IQ and chronological age were also investigated. We found an independent association between the homozygous DAT1 10/10 repeat genotype and RD that was not accounted for by the level of ADHD symptoms. This finding suggests that the DAT1 gene polymorphism may influence a common neural mechanism underlying both reading acquisition and ADHD symptoms.     

Conceptual blending of high-level features and data-driven salience computation in melodic generation

Conceptual Blending (CB) theory describes the cognitive mechanisms underlying the way humans process the emergence of new conceptual spaces by blending two input spaces. CB theory has been primarily used as a method for interpreting creative artefacts, while recently it has been utilised in the context of computational creativity for algorithmic invention of new concepts. Examples in the domain of music include the employment of CB interpretatively as a tool to explain musical semantic structures based on lyrics of songs or on the relations between body gestures and music structures. Recent work on generative applications of CB has shown that proper low-level representation of the input spaces allows the generation of consistent and sometimes surprising blends. However, blending high-level features (as discussed in the interpretative studies) of music explicitly, is hardly feasible with mere low-level representation of objects. Additionally, selecting features that are more salient in the context of two input spaces and relevant background knowledge and should, thus, be preserved and integrated in new interesting blends has not yet been tackled in a cognitively pertinent manner. The paper at hand proposes a novel approach to generating new material that allows blending high-level features by combining low-level structures, based on statistically computed salience values for each high-level feature extracted from data. The proposed framework is applied to a basic but, at the same time, complicated field of music, namely melodic generation. The examples presented herein allow an insightful examination of what the proposed approach does, revealing new possibilities and prospects.     

Demand response management system with discrete time window using supervised learning algorithm

Demand Response (DR) is a key attribute to enhance the operation of smart grid. Demand response improves the performance of the electric power systems and also deals with peak demand issues. Demand Response (DR) implementation for residential consumers is potentially accredited by Home Energy Management System (HEMS). This paper presents an algorithm for home energy management system to shift the schedulable loads in a residential home, that neglects consumer discomfort and minimizes electricity bill of energy consumption using Hourly-Time-Of-Use (HTOU) pricing scheme. Supervised learning algorithm is used in this paper to learn the usage patterns of consumers to allow schedulable appliances at a residential home to autonomously overcome consumer discomfort. Simulation results confirms that the proposed algorithm effectively decreases consumer electricity bill, decreases peak load demand and also avoids consumer discomfort.     

Analysis of structural equation model with ignorable missing continuous and polytomous data

The main purpose of this article is to develop a Bayesian approach for structural equation models with ignorable missing continuous and polytomous data. Joint Bayesian estimates of thresholds, structural parameters and latent factor scores are obtained simultaneously. The idea of data augmentation is used to solve the computational difficulties involved. In the posterior analysis, in addition to the real missing data, latent variables and latent continuous measurements underlying the polytomous data are treated as hypothetical missing data. An algorithm that embeds the Metropolis-Hastings algorithm within the Gibbs sampler is implemented to produce the Bayesian estimates. A goodness-of-fit statistic for testing the posited model is presented. It is shown that the proposed approach is not sensitive to prior distributions and can handle situations with a large number of missing patterns whose underlying sample sizes may be small. Computational efficiency of the proposed procedure is illustrated by simulation studies and a real example.The work described in this paper was fully supported by a grant from the Research Grants Council of the HKSAR (Project No. CUHK 4088/99H). The authors are greatly indebted to the Editor and anonymous reviewers for valuable comments in improving the paper; and also to D. E. Morisky and J.A. Stein for the use of their AIDS data set.     

Riemannian Newton and trust-region algorithms for analytic rotation in exploratory factor analysis

In exploratory factor analysis, latent factors and factor loadings are seldom interpretable until analytic rotation is performed. Typically, the rotation problem is solved by numerically searching for an element in the manifold of orthogonal or oblique rotation matrices such that the rotated factor loadings minimize a pre-specified complexity function. The widely used gradient projection (GP) algorithm, although simple to program and able to deal with both orthogonal and oblique rotation, is found to suffer from slow convergence when the number of manifest variables and/or the number of latent factors is large. The present work examines the effectiveness of two Riemannian second-order algorithms, which respectively generalize the well-established truncated Newton and trust-region strategies for unconstrained optimization in Euclidean spaces, in solving the rotation problem. When approaching a local minimum, the second-order algorithms usually converge superlinearly or even quadratically, better than first-order algorithms that only converge linearly. It is further observed in Monte Carlo studies that, compared to the GP algorithm, the Riemannian truncated Newton and trust-region algorithms require not only much fewer iterations but also much less processing time to meet the same convergence criterion, especially in the case of oblique rotation.     

基于可达阵的一种Q矩阵标定方法

Q矩阵标定是实施认知诊断评估的前提,已有Q矩阵修正方法并不太适合测验中已知属性向量的题目数较少的情形。根据拓展Q矩阵理论中可达阵R列与简化Q阵列存在布尔“或”关系,在一定认知假设下,率先提出可达阵R与简化Q阵的潜在反应列存在布尔“与”关系,并由此提出基于可达阵的Q矩阵标定方法。研究显示：在已知一个可达阵下,当可达阵项目的猜测或失误参数在.20以下且待标定项目的项目参数约在.30以下时,新方法所得Q矩阵元素返真率基本在.90以上,并且真实Q矩阵与估计Q矩阵下被试分类准确率差异很小;对于含5个属性的独立结构,新方法要求的随机样本的样本量较小;实证研究也印证了模拟研究的结论。新方法只需专家标定少量题目的Q矩阵,即已经标定的Q矩阵对应属性层级结构的可达阵。     

Genetic Virtue Program: An Unfeasible Neo-Pelagian Theodicy?

In his article, “Genetic Engineering, Virtue-First Enhancement, and Neo-Irenaean Theodicy,” Mark Walker has ventured farther into science more than most when it comes to exploring theodicy. After exposing the Achilles heel of the traditional free-will defense, Walker develops the Irenaean and Augustinian responses to the anthropic problem. Most importantly for this discussion, Walker proceeds to propose Genetic-First-Enhancement as part of his neo-Irenaean theodicy formulation. Overall, there are two major concerns I raised: the impossibility of a gradient morality in the presence of free will, and the scientific impossibility of Genetic Virtue Program. However, my claims are falsifiable if future genetic modifications do indeed improve morality. Before that is proven, I agree with Walker that, yes, we should play God, albeit, with his proposed virtue-first program.