Copying prompts with subsequent unprompted practice produced better learning of simple discriminations than feedback only of a correct response without subsequent practice. The Copy condition promoted faster acquisition of accurate performance for all subjects, and shorter response latencies and durations for 3 of 4 subjects. The data support the findings of Barbetta, Heron, and Heward, 1993 as well as Drevno, Kimball, Possi, Heward, Garner III, and Barbetta, 1994. The author proposes that response repertoires are most valuable if easily reacquired at times after original learning. Thus, reacquisition performance data are emphasized. The data suggest that discriminations acquired by copying prompts may result in useful repertoires if a practice procedure is used which facilitates transfer of stimulus control from a formal prompt to a naturally occurring stimulus.

This thesis describes practical learning systems able to model unknown nonlinear dynamic processes from their observed input-output behaviour. Local Model Networks use a number of simple, locally accurate models to represent a globally complex process, and provide a powerful, flexible framework for the integration of different model structures and learning algorithms. A major difficulty with Local Model Nets is the optimisation of the model structure. A novel Multi-Resolution Constructive (MRC) structure identification algorithm for local model networks is developed. The algorithm gradually adds to the model structure by searching for 'complexity' at ever decreasing scales of 'locality'. Reliable error estimates are useful during development and use of models. New methods are described which use the local basis function structure to provide interpolated state-dependent estimates of model accuracy. Active learning methods which automatically construct a training set for a given Local Model structure are developed, letting the training set grow in step with the model structure - the learning system 'explores' its data set looking for useful information. Local Learning methods developed in this work are explicitly linked to the local nature of the basis functions and provide a more computationally efficient method, more interpretable models and, due to the poor conditioning of the parameter estimation problem, often lead to an improvement in generalisation, compared to global optimisation methods. Important side-effects of normalisation of the basis functions are examined. A new hierarchical extension of Local Model Nets is presented: the Learning Hierarchy of Models (LHM), where local models can be sub-networks, leading to a tree-like hierarchy of softly interpolated local models. Constructive model structure identification algorithms are described, and the advantages of hierarchical 'divide-and-conquer' methods for modelling, especially in high dimensional spaces are discussed. The structures and algorithms are illustrated using several synthetic examples of nonlinear multivariable systems (dynamic and static), and applied to real world examples. Two nonlinear dynamic applications are described: predicting the strip thickness in an aluminium rolling mill from observed process data, and modelling robot actuator nonlinearities from measured data. The Local Model Nets reliably constructed models which provided the best results to date on the Rolling Mill application.

Thesis advisor: Rebecca Lowenhaupt / This qualitative case study examined the organizational learning mechanisms utilized by a district superintendent and their impact on principals’ learning. Examining recent curriculum reform efforts, the study concentrated on a small sample of building principals within a mid-sized urban public school district. Grounded in both organizational and situated learning theories, the research focused on organizational learning mechanisms and the interplay created by their implementation through the analysis of interview data and documents. Findings highlighted how the superintendent interpreted and distributed information to principals. In addition, findings showed the impact that superintendent-initiated processes, behaviors, and structures had on principal learning. The study provided strong evidence that the superintendent under study took steps to create district structures to support organizational learning. Moreover, principal data showed the impact of these structures on principals’ perceived learning. / Thesis (EdD) — Boston College, 2016. / Submitted to: Boston College. Lynch School of Education. / Discipline: Educational Leadership and Higher Education.

Typescript. / Digitized by Kansas Correctional Industries

Multiple instance (MI) learning is a relatively new topic in machine learning. It is concerned with supervised learning but differs from normal supervised learning in two points: (1) it has multiple instances in an example (and there is only one instance in an example in standard supervised learning), and (2) only one class label is observable for all the instances in an example (whereas each instance has its own class label in normal supervised learning). In MI learning there is a common assumption regarding the relationship between the class label of an example and the ``unobservable'' class labels of the instances inside it. This assumption, which is called the ``MI assumption'' in this thesis, states that ``An example is positive if at least one of its instances is positive and negative otherwise''. In this thesis, we first categorize current MI methods into a new framework. According to our analysis, there are two main categories of MI methods, instance-based and metadata-based approaches. Then we propose a new assumption for MI learning, called the ``collective assumption''. Although this assumption has been used in some previous MI methods, it has never been explicitly stated,\footnote{As a matter of fact, for some of these methods, it is actually claimed that they use the standard MI assumption stated above.} and this is the first time that it is formally specified. Using this new assumption we develop new algorithms --- more specifically two instance-based and one metadata-based methods. All of these methods build probabilistic models and thus implement statistical learning algorithms. The exact generative models underlying these methods are explicitly stated and illustrated so that one may clearly understand the situations to which they can best be applied. The empirical results presented in this thesis show that they are competitive on standard benchmark datasets. Finally, we explore some practical applications of MI learning, both existing and new ones. This thesis makes three contributions: a new framework for MI learning, new MI methods based on this framework and experimental results for new applications of MI learning.

Motivated by various challenging real-world applications, such as drug activity prediction and image retrieval, multi-instance (MI) learning has attracted considerable interest in recent years. Compared with standard supervised learning, the MI learning task is more difficult as the label information of each training example is incomplete. Many MI algorithms have been proposed. Some of them are specifically designed for MI problems whereas others have been upgraded or adapted from standard single-instance learning algorithms. Most algorithms have been evaluated on only one or two benchmark datasets, and there is a lack of systematic comparisons of MI learning algorithms. This thesis presents a comprehensive study of MI learning algorithms that aims to compare their performance and find a suitable way to properly address different MI problems. First, it briefly reviews the history of research on MI learning. Then it discusses five general classes of MI approaches that cover a total of 16 MI algorithms. After that, it presents empirical results for these algorithms that were obtained from 15 datasets which involve five different real-world application domains. Finally, some conclusions are drawn from these results: (1) applying suitable standard single-instance learners to MI problems can often generate the best result on the datasets that were tested, (2) algorithms exploiting the standard asymmetric MI assumption do not show significant advantages over approaches using the so-called collective assumption, and (3) different MI approaches are suitable for different application domains, and no MI algorithm works best on all MI problems.

Thesis (M. Ed.)--University of Hong Kong, 2006. / Title proper from title frame. Also available in printed format.

Recent work has shown that AdaBoost can be viewed as an algorithm that maximizes the margin on the training data via functional gradient descent. Under this interpretation, the weight computed by AdaBoost, for each hypothesis generated, can be viewed as a step size parameter in a gradient descent search. Friedman has suggested that shrinking these step sizes could produce improved generalization and reduce overfitting. In a series of experiments, he showed that very small step sizes did indeed reduce overfitting and improve generalization for three variants of Gradient_Boost, his generic functional gradient descent algorithm. For this report, we tested whether reduced learning rates can also improve generalization in AdaBoost. We tested AdaBoost (applied to C4.5 decision trees) with reduced learning rates on 28 benchmark datasets. The results show that reduced learning rates provide no statistically significant improvement on these datasets. We conclude that reduced learning rates cannot be recommended for use with boosted decision trees on datasets similar to these benchmark datasets. / Graduation date: 2002

This is a case study, within the field of Education and Human Resource Development. The subject is handover in a non-profit organization. The organization studied is the Uppsala Student Union (US). US is a politically run Non-profit organization (NGO), with the objective to work for better study- and living conditions of the 35 000 students at Uppsala University, Sweden, who are its members. Four people active within US have been interviewed, and the empiric material has been analyzed mainly based on the theories of Organizational Learning and Continuity Management. Some other theories are presented as well, as an orientation with relation to handover in organizations and organizational development. The questions asked are: 1. How is transfer of knowledge perceived in US – as a significant problem, a small problem or no problem at all? 2. If transfer of knowledge is perceived as a problem, what are thought to be the causes? 3. In US, as a NGO, how is handover managed? Which problems arise with respect to handover? 4. What else of interest and relevance can be found? The answers are: 1. A small problem. Transfer of knowledge is much thought of, but there are problems which are viewed as more important. 2. The causes for problems with handover are mainly referred to a heavy workload for the actives, high turnover and insufficient handover routines. 3. Routines for handover is a well integrated part of the work at US. And the conditions in terms of resources are good compared to other student unions. Some problems still arise, and a selection of these are presented in the study. 4. Additional findings have been defined under the following headlines: Representation on Boards – an area for improvement Changing the roles On Actives-mentality (Swe. föreningsmänniskor) Effective policy making Students as actives

This study aims to explore the effects on junior high school students¡¦ learning attitudes as well as learning performance after using 3C products. The study was conducted by means of questionnaire survey with self-editted ¡§Questionnaire on Learning Attitudes and Learning Performance¡¨. 600 junior high school students were randomly sampled from 25 public junior high schools in Kaohsing County and Kaohsiung City. The collected data was analyzed by statistical methods, including T-test, Chi-Square test, One-Way ANOVA, Point-Biserial Correlation , Person¡¦s Correlation and Multiple Regression. Based on the analyzed results, the followings were concluded: 1. The students of different background have different using function on 3C products such as cellphones, console game, digital cameras and computers. 2. The students of different background have different using frequency on 3C products such as cellphones, console game, digital cameras . 3. The effects occurred on students¡¦ attitudes after using 3C products. 4. Students¡¦ performance was affected by their attitudes after using 3C products. 5. The predictability on learning performance of Skill Level can even reach to 53.1%, followed by 1.6% of Emotion Level.