Model Selection for Unsupervised Learning of Visual Context

This study addresses the problem of choosing the most suitable probabilistic model selection criterion for unsupervised learning of visual context of a dynamic scene using mixture models. A rectified Bayesian Information Criterion (BICr) and a Completed Likelihood Akaike’s Information Criterion (CL-AIC) are formulated to estimate the optimal model order (complexity) for a given visual scene. Both criteria are designed to overcome poor model selection by existing popular criteria when the data sample size varies from small to large and the true mixture distribution kernel functions differ from the assumed ones. Extensive experiments on learning visual context for dynamic scene modelling are carried out to demonstrate the effectiveness of BICr and CL-AIC, compared to that of existing popular model selection criteria including BIC, AIC and Integrated Completed Likelihood (ICL). Our study suggests that for learning visual context using a mixture model, BICr is the most appropriate criterion given sparse data, while CL-AIC should be chosen given moderate or large data sample sizes.     

Performance discrepancies on the California Verbal Learning Test-Second Edition (CVLT-II) in the standardization sample.

The standardization data for the California Verbal Learning Test-Second Edition (CVLT-II; D. C. Delis, J. H. Kramer, E. Kaplan, & B. A. Ober, 2000) were used to evaluate the base rate of 6 specific discrepancies between various key variables. The results indicated that CVLT-II performance discrepancies should equal or exceed 1 or 1.5 z score points (depending on the individual comparison) in the hypothesized direction to be considered potentially unusual. However, because about 1 in 3 persons in the standardization sample displayed at least 1 such large discrepancy, it is concluded that these base rates should be viewed only as a starting point for interpretation. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Quantitative pharmacophore models with inductive logic programming

Three-dimensional models, or pharmacophores, describing Euclidean constraints on the location on small molecules of functional groups (like hydrophobic groups, hydrogen acceptors and donors, etc.), are often used in drug design to describe the medicinal activity of potential drugs (or ‘ligands’). This medicinal activity is produced by interaction of the functional groups on the ligand with a binding site on a target protein. In identifying structure-activity relations of this kind there are three principal issues: (1) It is often difficult to “align” the ligands in order to identify common structural properties that may be responsible for activity; (2) Ligands in solution can adopt different shapes (or `conformations’) arising from torsional rotations about bonds. The 3-D molecular substructure is typically sought on one or more low-energy conformers; and (3) Pharmacophore models must, ideally, predict medicinal activity on some quantitative scale. It has been shown that the logical representation adopted by Inductive Logic Programming (ILP) naturally resolves many of the difficulties associated with the alignment and multi-conformation issues. However, the predictions of models constructed by ILP have hitherto only been nominal, predicting medicinal activity to be present or absent. In this paper, we investigate the construction of two kinds of quantitative pharmacophoric models with ILP: (a) Models that predict the probability that a ligand is “active”; and (b) Models that predict the actual medicinal activity of a ligand. Quantitative predictions are obtained by the utilising the following statistical procedures as background knowledge: logistic regression and naive Bayes, for probability prediction; linear and kernel regression, for activity prediction. The multi-conformation issue and, more generally, the relational representation used by ILP results in some special difficulties in the use of any statistical procedure. We present the principal issues and some solutions. Specifically, using data on the inhibition of the protease Thermolysin, we demonstrate that it is possible for an ILP program to construct good quantitative structure-activity models. We also comment on the relationship of this work to other recent developments in statistical relational learning. Editors: Tamás Horváth and Akihiro Yamamoto     

Implicit Motor Learning in Parkinson's Disease.

Objective: This study examined whether disruption of performance is moderated in Parkinson's disease (PD) patients who acquire their motor behaviors in an implicit manner. Method: Twenty-seven patients with PD learned a hammering task in errorless (implicit) or errorful (explicit) conditions and were tested for robustness of motor performance under a secondary task load, which required them to continuously count backward as they performed the hammering task. Results: Patients in the errorless (implicit) motor learning condition exhibited robustness to secondary task loading, whereas patients in the errorful (explicit) motor learning condition did not. Conclusions: Implicit motor learning techniques should be considered by PD rehabilitation specialists in cases in which existing disruption to movements is exacerbated by conscious control. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

NMDA Receptor Modulation of Incidental Learning in Pavlovian Context Conditioning.

Rats exposed to a footshock show conditional fear when reexposed to the shock context. Immediate presentation of shock after placement in the context significantly reduces this fear. Preexposure to the context in the absence of shock, coupled with a minimum preshock interval during training, overcomes this immediate shock deficit. Because rats learn about the context during preexposure and express that learning after being reinforced, the context preexposure effect is an aversive analogue of latent learning. The authors examined the effect of the N-methyl-D-aspartate (NMDA) receptor antagonist D,L-2-amino-5-phosphovalerate (APV) on the facilitatory effect of context preexposure. Rats were preexposed to a chamber after APV administration. The next day they were placed in the same chamber without drug and received shock 35 s later. APV blocked the facilitatory effect of preexposure. Therefore NMDA receptors are important for contextual latent learning. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

基于支持向量机的多类分类研究

现今流行的分类方法的重要基础是传统的统计学，前提是要有足够的样本，当样本数目有限时容易出现过学习的问题，导致分类效果不理想。引入支持向量机方法，它基于统计学习理论，采用了结构风险最小化原则代替经验风险最小化原则，较好的解决了小样本学习的问题；又由于采用了核函数思想，把非线性空间的问题转换到线性空间，降低了算法的复杂度。对其相关内容包括优化算法及多类分类问题的解决进行了研究，最后用一个实例说明了该方法的可行性和有效性。     

An intelligent web-page classifier with fair feature-subset selection

The explosion of on-line information has given rise to many manually constructed topic hierarchies (such as Yahoo!!). But with the current growth rate in the amount of information, manual classification in topic hierarchies results in an immense information bottleneck. Therefore, developing an automatic classifier is an urgent need. However, classifiers suffer from enormous dimensionality, since the dimensionality is determined by the number of distinct keywords in a document corpus. More seriously, most classifiers are either working slowly or they are constructed subjectively without any learning ability. In this paper, we address these problems with a fair feature-subset selection (FFSS) algorithm and an adaptive fuzzy learning network (AFLN) for classification. The FFSS algorithm is used to reduce the enormous dimensionality. It not only gives fair treatment to each category but also has ability to identify useful features, including both positive and negative features. On the other hand, the AFLN provides extremely fast learning ability to model the uncertain behavior for classification so as to correct the fuzzy matrix automatically. Experimental results show that both FFSS algorithm and the AFLN lead to a significant improvement in document classification, compared to alternative approaches.     

Preschool Competency in Context: An Investigation of the Unique Contribution of Child Competencies to Early Academic Success.

The present concurrent study combined developmental and ecological considerations to examine the unique contribution of multiple preschool competencies to an indicator of early academic success. Participants included 195 Head Start children from 32 classrooms representative of a large, urban Head Start program. Dimensional (variable-centered) analyses revealed 3 distinct classroom competency dimensions (i.e., General Classroom Competencies, Specific Approaches to Learning, and Interpersonal Classroom Behavioral Problems). The first 2 of these dimensions were found to be uniquely associated with early academic success. Findings from typological (person-centered) analyses supported the dimensional findings. Typological analyses revealed 7 profiles of classroom competency distinguished by high scores on the dimensions of General Competencies and Approaches to Learning, and these profiles were found to relate differentially to the indicator of early academic success. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

一种T-S型动态模糊推理模型的建模算法研究

以一种进化聚类算法（ECM）为基础,提出了一种新的T-S型动态模糊推理模型的建模算法.以往许多神经模糊模型都不适用于自适应在线学习,而文章模型能够实时地调整模糊规则库及规则参数,具有较强的在线学习能力.仿真结果表明,该方法是有效的.     

A Revised Methodology for Research on Metamemory: Pre-judgment Recall And Monitoring (PRAM).

A revised methodology is described for research on metacognitive monitoring, especially judgments of learning (JOLs), to investigate psychological processing that previously has been only hypothetical and unobservable. During data collection a new stage of recall occurs just prior to the JOL, so that during data analysis the items can be partitioned into subcategories to measure the degree of JOL accuracy in ways that are more analytic than was previously possible. A weighted-average combinatorial rule allows the component measures of JOL accuracy to be combined into the usual overall measure of metacognitive accuracy. An example using the revised methodology offers a new explanation for the delayed-JOL effect, in which delayed JOLs are more accurate than immediate JOLs for predicting recall. (PsycINFO Database Record (c) 2010 APA, all rights reserved)