Thesis advisor: Martin Scanlan / District improvement research highlights the importance of central office leaders and building principals in reform efforts. A growing body of literature emphasizes the importance of the relationship between central office leaders and principals in the context of turnaround school reform. Drawing on research from Honig (2012), the assistance relationship consistently surfaced as a key element of success when selecting and implementing learning time as a turnaround measure. This study aims to explore the assistance relationship between central office leaders and principals in the selection and implementation of learning time. This qualitative study draws upon observations and document review to answer the research questions, and is one strand of a larger research project studying how central office leaders support principals through an assistance relationship in a district in receivership. This individual strand concludes that the selection and implementation of learning time opportunities, without consideration to the principals and without the assistance of central office leaders, can’t happen. This strand advances our understanding of how an effective assistance relationship can work in a district in receivership. I concluded that central office leaders and principals interviewed selected and implemented like learning time opportunities. Future researchers may continue to enrich this growing body of literature by examining these findings and testing all or some of the five key practices in a like district. The results, implications for districts in receivership, and future research are discussed. This strand’s findings can serve as a companion for central office leaders who are working with principals to select and implement learning time opportunities in order to improving achievement levels in underperforming schools and districts across our country. / Thesis (EdD) — Boston College, 2018. / Submitted to: Boston College. Lynch School of Education. / Discipline: Educational Leadership and Higher Education.

The aim of the thesis was to create a tool for time-series prediction based on deep learning. The first part of the work is a brief description of deep learning and its comparison to classical machine learning. In the next section contains brief analysis of some tools, that are already used for time-series forecasting. The last part is focused on the analysis of the problem as well as on the actual creation of the program.

Time series forecasting has become a common problem in day-to-day applications and various machine learning algorithms have been developed to tackle this task. Finding the model that performs the best forecasting on a given dataset can be time consuming as multiple algorithms and hyperparameter configurations must be examined to find the best model. This problem can be solved using automated machine learning, an approach that automates all steps required for developing a machine learning algorithm including finding the best algorithm and hyperparameter configuration. This study develops and builds an automated machine learning pipeline focused on finding the best forecasting model for a given dataset. This includes choosing different forecasting algorithms to cover a wide range of tasks and identifying the best method to find the best model in these algorithms. Lastly, the final pipeline will then be tested on a variety of datasets to evaluate the performance on time series data with different characteristics.:Abstract List of Figures List of Tables List of Abbreviations List of Symbols 1. Introduction 2. Theoretical Background 2.1. Machine Learning 2.2. Automated Machine Learning 2.3. Hyperparameter Optimization 2.3.1. Model-Free Methods 2.3.2. Bayesian Optimization 3. Time Series Forecasting Algorithms 3.1. Time Series Data 3.2. Baselines 3.2.1. Naive Forecast 3.2.2. Moving Average 3.3. Linear Regression 3.4. Autoregression 3.5. SARIMAX 3.6. XGBoost 3.7. LSTM Neural Network 4. Automated Machine Learning Pipeline 4.1. Data Preparation 4.2. Model Selection 4.3. Hyperparameter Optimization Method 4.3.1. Sequential Model-Based Algorithm Configuration 4.3.2. Tree-structured Parzen Estimator 4.3.3. Comparison of Bayesian Optimization Hyperparameter Optimization Methods 4.4. Pipeline Structure 5. Testing on external Datasets 5.1. Beijing PM2.5 Pollution 5.2. Perrin Freres Monthly Champagne Sales 6. Testing on internal Datasets 6.1. Deutsche Telekom Call Count 6.1.1. Comparison of Bayesian Optimization and Random Search 6.2. Deutsche Telekom Call Setup Time 7. Conclusion Bibliography A. Details Search Space B. Pipeline Results - Predictions C. Pipeline Results - Configurations D. Pipeline Results - Experiment Details E. Deutsche Telekom Data Usage Permissions

The past twenty years have seen a successful formalization of the idea that perception is a form of probabilistic inference. Bayesian Decision Theory (BDT) provides a neat mathematical framework for describing how an ideal observer and actor should interpret incoming sensory stimuli and act in the face of uncertainty. The predictions of BDT, however, crucially depend on the observer’s internal models, represented in the Bayesian framework by priors, likelihoods, and the loss function. Arguably, only in the simplest scenarios (e.g., with a few Gaussian variables) we can expect a real observer’s internal representations to perfectly match the true statistics of the task at hand, and to conform to exact Bayesian computations, but how humans systematically deviate from BDT in more complex cases is yet to be understood. In this thesis we theoretically and experimentally investigate how people represent and perform probabilistic inference with complex (beyond Gaussian) one-dimensional distributions of stimuli in the context of sensorimotor decision making. The goal is to reconstruct the observers’ internal representations and details of their decision-making process from the behavioural data – by employing Bayesian inference to uncover properties of a system, the ideal observer, that is believed to perform Bayesian inference itself. This “inverse problem” is not unique: in principle, distinct Bayesian observer models can produce very similar behaviours. We circumvented this issue by means of experimental constraints and independent validation of the results. To understand how people represent complex distributions of stimuli in the specific domain of time perception, we conducted a series of psychophysical experiments where participants were asked to reproduce the time interval between a mouse click and a flash, drawn from a session-dependent distribution of intervals. We found that participants could learn smooth approximations of the non-Gaussian experimental distributions, but seemed to have trouble with learning some complex statistical features such as bimodality. To investigate whether this difficulty arose from learning complex distributions or computing with them, we conducted a target estimation experiment in which “priors” where explicitly displayed on screen and therefore did not need to be learnt. Lack of difference in performance between the Gaussian and bimodal conditions in this task suggests that acquiring a bimodal prior, rather than computing with it, is the major difficulty. Model comparison on a large number of Bayesian observer models, representing different assumptions about the noise sources and details of the decision process, revealed a further source of variability in decision making that was modelled as a “stochastic posterior”. Finally, prompted by a secondary finding of the previous experiment, we tested the effect of decision uncertainty on the capacity of the participants to correct for added perturbations in the visual feedback in a centre of mass estimation task. Participants almost completely compensated for the injected error in low uncertainty trials, but only partially so in the high uncertainty ones, even when allowed sufficient time to adjust their response. Surprisingly, though, their overall performance was not significantly affected. This finding is consistent with the behaviour of a Bayesian observer with an additional term in the loss function that represents “effort” – a component of optimal control usually thought to be negligible in sensorimotor estimation tasks. Together, these studies provide new insight into the capacity and limitations people have in learning and performing probabilistic inference with distributions beyond Gaussian. This work also introduces several tools and techniques that can help in the systematic exploration of suboptimal behaviour. Developing a language to describe suboptimality, mismatching representations and approximate inference, as opposed to optimality and exact inference, is a fundamental step to link behavioural studies to actual neural computations.

Academic learning time (ALT) has long had the theoretical underpinnings sufficient to claim a causal relationship with academic achievement, but to this point empirical evidence has been lacking. This dearth of evidence has existed primarily due to difficulties associated with operationalizing ALT in traditional educational settings. Recent advancements in computer-based instruction provide an unprecedented opportunity to model ALT and to test the underlying theory. A widely-used computer-based early-reading curriculum was operationalized using Berliner's model of ALT (Berliner, 1991). This curriculum was then mapped to a computer-based assessment to determine an appropriate method of quantifying early-reading ALT. Software limitations required that ALT be quantified as a summative measure. Data were collected from 1,347 prekindergarteners and were analyzed using a dose-response approach that associated usage of the curriculum with a generalized variable of early-reading achievement. Gains across four early-reading skills were demonstrated via linear regression to be predicted by minutes of usage (Adj. R2 = .078). A sample optimized to test the hypothesis showed a stronger correlation (Adj. R2 = .096). Time spent using the Free Play version of the curriculum did not uniquely predict additional variance. Similarly, gains on reading skills that were not taught explicitly by the curriculum were not predicted by overall usage. These three results were interpreted as supporting the ALT learning model. Post-hoc analyses were performed on curriculum-usage compliance and on within-curriculum progress, both of which were statistically significant when added to the basic dose-response model. Multiple exploratory best-fit models were constructed. The strongest accounted for just under 20% of the overall variance (Adj. R2 = .186). Effect sizes were in the medium-to-large range for the entire sample (D = 0.71) with significant improvement for the optimized sample (D = 1.26). Children in the optimized sample who used the program over 20% more than recommended had even stronger gains (D = 1.67). The ability to remotely and accurately quantify interaction with a computer-based curriculum and assessment in the home defines a new vista in ALT research.

Effective mental workload measurement is critical because mental workload significantly affects human performance. A non-invasive and objective workload measurement tool is needed to overcome limitations of current mental workload measures. Further, training/learning increases mental workload during skill or knowledge acquisition, followed by a decreased mental workload, though sufficient training times are unknown. The objectives of this study were to: (1) investigate the efficacy of using thermography as a non-contact physiological measure to quantify mental workload, (2) quantify and describe the relationship between mental workload and learning/training, and, (3) introduce a method to determine a sufficient training time and an optimal human performance level for a novel task by using thermography. Three studies were conducted to address these objectives. The first study investigated the efficacy of using thermography to quantity the relationship between mental workload and facial temperature changes while learning an alpha-numeric task. Thermography measured and quantified the mental workload level successfully. Strong and significant correlations were found among thermography, performance, and subjective workload measures (MCH and SWAT ratings). The second study investigated the utility of using a psychophysical approach to determine workload levels that maximize performance on a cognitive task. The second study consisted of an adjustment session (participants adjusted their own workload levels) and work session (participants worked at the chosen workload level). Participants were found to fall into two performance groups (low and high performers by accuracy rate) and results were significantly different. Thermography demonstrated whether both group found their optimal workload level. The last study investigated efficacy of using thermography to quantify mental workload level in a complex training/learning environment. Experienced drivers’ performance data was used as criteria to indicate whether novice drivers mastered the driving skills. Strong and significant correlations were found among thermography, subjective workload measures, and performance measures in novice drivers. This study verified that thermography is a reliable and valid way to measure workload as a non-invasive and objective method. Also, thermography provided more practical results than subjective workload measures for simple and complex cognitive tasks. Thermography showed the capability to identify a sufficient training time for simple or complex cognitive tasks.

No description available.

Reading has long been acknowledged to be a critical skill that is best acquired early in life. According to the most recent National Assessment of Educational Progress (NAEP) reports, American public school children continue to struggle to master the basics of reading. Although federal funding in real dollars has increased consistently over time, reading scores have not followed suit, suggesting that fiscal resources have not been applied successfully to the variables that are directly related to reading acquisition and achievement. The current state of affairs therefore suggests the need for identifying a fiscally-targetable, instructionally-relevant variable with a direct, causal relationship to early-reading achievement. One way to determine whether such a relationship exists between two variables is by means of dose-response methodology. Although this methodology has not been broadly implemented in educational research, it is attractive because it allows for the formal characterization and comparison of cause-effect relationships, and may also inform practice in readily implementable ways. Researchers have noted that time spent learning (TSL), and in particular academic learning time (ALT), is a promising candidate for a dose-response relationship with student achievement in early reading. Although ALT holds promise, there have traditionally been significant difficulties in operationalizing and quantifying it. The growing prevalence of academic software in the American public classroom holds promise for overcoming these challenges and provides an opportunity to test the hypothesis that there is a dose-response relationship between ALT and student achievement in early reading.

Over the past few years many changes have taken place in the content and presentation of teacher education programmes in South Africa due to the paradigm shift from teaching to learning. As a result, the primary learning environment for undergraduate students, the fairly passive lecture-discussion format where teacher educators talk and most students listen, is contrary to almost every principle of an optimal student learning setting. The current view in teacher education is that teacher educators should create learner-centred and learner-controlled environments where student learning and success determine the boundary. The idea of focusing on learning rather than teaching requires that teacher educators rethink their role and the role of students in the teaching and learning process. When focussing on learning rather than teaching, teacher educators must challenge their basic assumptions about how people learn and what the roles of teacher educators should be. It may be necessary to unlearn previously acquired teaching habits, and rethink the role of assessment and feedback in learning. Meaningful, formative assessment can play a key role in shifting to a learner-centred approach because it provides important information to both students and teacher educators at all stages of the learning process. To achieve this, it is essential that teacher educators do not simply add assessment as an extra to an existing, non-interactive scheme of work, but that they integrate assessment effectively and efficiently with their instruction. This requires a major shift in how assessment is planned and integrated and a working framework for integrating assessment with instruction can be most valuable to teacher educators. The purpose of this study was to: Determine the nature and scope of ESL teacher educators' tasks, within a Faculty of Education Sciences, at a tertiary institution. Determine the extent to which ESL teacher educators are implementing a learner-centred approach to teaching and learning. Identify the factors, if any, that impede the transition to a learner-centred approach to teaching and learning. Provide recommendations to facilitate the implementation of a learner-centred approach to teaching and learning. Determine how, when and how often ESL teacher educators are currently conducting assessment. Identify possible shortcomings of the existing assessment system of ESL teacher educators. Provide a framework for implementing assessment within a learner-centred approach to teaching and learning. A one-shot cross-sectional survey design was used in this study. The participants included all the teacher educators (N=5) within the Subject Group English in the Faculty of Education Sciences .at the Potchefstroom University. Three data collection techniques were used in this study, namely a questionnaire, semi-structured interviews and classroom observations. The purpose was to triangulate the data in order to get as complete a picture as possible of the extent to which the teacher educators' teaching and learning ~racticesre flected a focus on learner-centredness. The results of the study can be summarised as follows: Descriptive statistics (means and percentages) were used to analyse the data. The data collected during the interviews were reported as narratives. The results indicated that the teacher educators in this study spent a significant percentage of their time on preparation for class meetings and assessment. Each teacher educator taught for the full twelve weeks of each semester and, therefore, did not have one week free of teaching the entire year. Although the teacher educators embraced some learner-centred methods such as group work and interactive class discussions, they still assumed most of the responsibility for the learning processes and classroom behaviour of the students. They mainly focused on what to present in the contact sessions and spent time organizing presentations of information rather than developing materials to facilitate learning. The teacher educators often reverted to more familiar, traditional approaches and emphasized the following issues as affecting the effective and efficient transition to learner-centredness: curriculum coverage and lack of time, lack of proper training, size of student groups, other teacher educators' cynical attitudes and students' attitudes towards learning. The teacher educators made use of a variety of assessment methods and assessed students continuously, but these assessments were not used for promoting student learning, but rather for grading purposes. Students received traditional feedback such as grades, marks and scores, but they seldom received feedback on what they did wrong and how they could rectify it. Overall, it was assessment of learning and not assessment for learning. A major factor impeding the implementation of a learner-centred assessment approach was the demand formative assessment methods placed on the professional time of the teacher educators. In order to utilise time effectively and integrate assessment with the instructional design, teacher educators expressed the need for a workable framework to assist them in planning their assessment practices. / Thesis (M.Ed.)--North-West University, Potchefstroom Campus, 2004.

Over the past few years many changes have taken place in the content and presentation of teacher education programmes in South Africa due to the paradigm shift from teaching to learning. As a result, the primary learning environment for undergraduate students, the fairly passive lecture-discussion format where teacher educators talk and most students listen, is contrary to almost every principle of an optimal student learning setting. The current view in teacher education is that teacher educators should create learner-centred and learner-controlled environments where student learning and success determine the boundary. The idea of focusing on learning rather than teaching requires that teacher educators rethink their role and the role of students in the teaching and learning process. When focussing on learning rather than teaching, teacher educators must challenge their basic assumptions about how people learn and what the roles of teacher educators should be. It may be necessary to unlearn previously acquired teaching habits, and rethink the role of assessment and feedback in learning. Meaningful, formative assessment can play a key role in shifting to a learner-centred approach because it provides important information to both students and teacher educators at all stages of the learning process. To achieve this, it is essential that teacher educators do not simply add assessment as an extra to an existing, non-interactive scheme of work, but that they integrate assessment effectively and efficiently with their instruction. This requires a major shift in how assessment is planned and integrated and a working framework for integrating assessment with instruction can be most valuable to teacher educators. The purpose of this study was to: Determine the nature and scope of ESL teacher educators' tasks, within a Faculty of Education Sciences, at a tertiary institution. Determine the extent to which ESL teacher educators are implementing a learner-centred approach to teaching and learning. Identify the factors, if any, that impede the transition to a learner-centred approach to teaching and learning. Provide recommendations to facilitate the implementation of a learner-centred approach to teaching and learning. Determine how, when and how often ESL teacher educators are currently conducting assessment. Identify possible shortcomings of the existing assessment system of ESL teacher educators. Provide a framework for implementing assessment within a learner-centred approach to teaching and learning. A one-shot cross-sectional survey design was used in this study. The participants included all the teacher educators (N=5) within the Subject Group English in the Faculty of Education Sciences .at the Potchefstroom University. Three data collection techniques were used in this study, namely a questionnaire, semi-structured interviews and classroom observations. The purpose was to triangulate the data in order to get as complete a picture as possible of the extent to which the teacher educators' teaching and learning ~racticesre flected a focus on learner-centredness. The results of the study can be summarised as follows: Descriptive statistics (means and percentages) were used to analyse the data. The data collected during the interviews were reported as narratives. The results indicated that the teacher educators in this study spent a significant percentage of their time on preparation for class meetings and assessment. Each teacher educator taught for the full twelve weeks of each semester and, therefore, did not have one week free of teaching the entire year. Although the teacher educators embraced some learner-centred methods such as group work and interactive class discussions, they still assumed most of the responsibility for the learning processes and classroom behaviour of the students. They mainly focused on what to present in the contact sessions and spent time organizing presentations of information rather than developing materials to facilitate learning. The teacher educators often reverted to more familiar, traditional approaches and emphasized the following issues as affecting the effective and efficient transition to learner-centredness: curriculum coverage and lack of time, lack of proper training, size of student groups, other teacher educators' cynical attitudes and students' attitudes towards learning. The teacher educators made use of a variety of assessment methods and assessed students continuously, but these assessments were not used for promoting student learning, but rather for grading purposes. Students received traditional feedback such as grades, marks and scores, but they seldom received feedback on what they did wrong and how they could rectify it. Overall, it was assessment of learning and not assessment for learning. A major factor impeding the implementation of a learner-centred assessment approach was the demand formative assessment methods placed on the professional time of the teacher educators. In order to utilise time effectively and integrate assessment with the instructional design, teacher educators expressed the need for a workable framework to assist them in planning their assessment practices. / Thesis (M.Ed.)--North-West University, Potchefstroom Campus, 2004.