Since reinforcement learning algorithms have to fully solve a task in order to evaluate a set of hyperparameter values, conventional hyperparameter tuning methods can be highly sample inefficient and computationally expensive. Many widely used reinforcement learning architectures originate from scientific papers which include optimal hyperparameter values in the publications themselves, but do not indicate how the hyperparameter values were found. To address the issues related to hyperparameter tuning, three different experiments were investigated. In the first two experiments, Bayesian Optimisation and random search are compared. In the third and final experiment, the hyperparameter values found in second experiment are used to solve a more difficult reinforcement learning task, effectively performing hyperparameter transfer learning (later referred to as meta-transfer learning). The results from experiment 1 showed that there are certain scenarios in which Bayesian Optimisation outperforms random search for hyperparameter tuning, while the results of experiment 2 show that as more hyperparameters are simultaneously tuned, Bayesian Optimisation consistently finds better hyperparameter values than random search. However, BO took more than twice the amount of time to find these hyperparameter values than random search. Results from the third and final experiment indicate that hyperparameter values learned while tuning hyperparameters for a relatively easy to solve reinforcement learning task (Task A), can be used to solve a more complex task (Task B). With the available computing power for this thesis, hyperparameter optimisation was possible on the tasks in experiment 1 and experiment 2. This was not possible on the task in experiment 3, due to limited computing resources and the increased complexity of the reinforcement learning task in experiment 3, making the transfer of hyperparameters from one task (Task A) to the more difficult task (Task B) highly beneficial for solving the more computationally expensive task. The purpose of this work is to explore the effectiveness of Bayesian Optimisation as a hyperparameter tuning algorithm on the reinforcement learning algorithm NEAT's hyperparemters. An additional goal of this work is the experimental use of hyperparameter value transfer between reinforcement learning tasks, referred to in this work as Meta-Transfer Learning. This is introduced and discussed in greater detail in the Introduction chapter. All code used for this work is available in the repository: • https://github.com/maaxnaax/MSc_code

Indiana University-Purdue University Indianapolis (IUPUI) / Heart failure is a terminal disease with an unpredictable trajectory. Family members of patients with heart failure are often called upon to make decisions about treatment and end of life care, sometimes with little guidance as to the patients’ wishes. Advance care planning (ACP) is an ongoing process by which patients make decisions about their future healthcare. Only about one-third of patients with heart failure have participated in ACP, which is a similar percentage to the overall population. Despite increased focus on ACP and interventions to improve it, the rates of ACP in the population remain relatively unchanged. There is a need to develop interventions that are targeted based on patient engagement in the process rather than the existing broad-based interventions. The purpose of this dissertation study is to examine the relationship between the American Heart Association stage of heart failure and readiness to engage in advance care planning. The study consisted of mailed surveys that consisted of demographic questionnaires and the Advance Care Planning Engagement Survey. Engagement was analyzed in relation to heart failure stage, heart failure class, comorbidities, perception of health status, recent hospitalizations, making healthcare decisions for others, and demographic variables. The results demonstrated that although there was no significant association between heart failure stage or class and engagement in advance care planning, there were significant associations between medical comorbidities and advance care planning engagement. Other significantly associated participant characteristics included age, gender, education, ethnicity, and income. Findings suggest that people with multiple comorbid conditions will be more likely to be ready to engage in ACP than those with fewer health conditions. The results from this study will contribute to the development of strategies to improve advance care planning that are targeted based on engagement level.

No description available.

No description available.

Thesis (Ph. D.)--University of Virginia, 1997. / Includes bibliographical references (105-117). Also available online through Digital Dissertations.

Includes bibliographical references. / In the post or neo-colonial era, the question of fair and equitable treatment of indigenous peoples remains a subject of international political and legal discourse. Efforts have been made to study ways of promoting and protecting indigenous rights and to develop international norms for the protection of these rights. These efforts have sprung forth a plethora of questions; these questions include 'who qualifies as indigenous peoples?' and 'what rights do they enjoy under international law.' This thesis takes a cursory look at the conceptual underpinnings of indigenous peoples and specifically evaluates their right to culture in the parlance of international law.

Numerous studies highlighted sharp declines in abundance of red oak species (Quercus spp., Section Erythrobalanus) in the southeastern United States. Red oaks are major components of bottomland forests, provide important ecological services, and are a critical source of hard mast for wildlife and high-value timber (Oliver et al. 2005). Bottomland hardwoods are usually managed with natural regeneration, and maintaining a component of red oak can challenge forest managers, given sporadic acorn production (masting behavior), and lack of advance regeneration establishment prior to disturbance. This study investigated the development of hardwood advance regeneration in relation to understory light availability and stand structure in mature closed canopy stands following silvicultural treatments. Improved understanding of red oak natural regeneration can better clarify any relationship between seedling abundance, available understory light, and residual basal area. Results will aid in selection of appropriate management techniques to sustain dominance of red oaks within bottomland hardwood forests.

本研究主要探討高工學生運用前置組織(Advance Organizer)於錄影帶教學輔助上所產生的聽力理解成效。 研究對象為台灣南部一國立高工化工、機械、室內設計科等四個班145位學生。使用全民英檢兩屆考題為前後測，並以前測成績區分高、中、低分群。實驗組給予聽前單字提攜與問題引導策略教學並佐以Family Album, U. S. A和A+English聽力教材，而控制組則只有錄影帶教材而無任何聽前引導。實驗研究共進行十六週，實驗前給予兩組學生學習背景問卷調查，實驗後由實驗組學生填寫學習回顧問卷。經由測驗、問卷兩方法得知，受試學生對於此前置引導錄影帶教學輔助在聽力理解上呈現顯着進步，其中又以高分群的成效最佳。 最後，由此研究結果提出工職英語聽力教學之具體建議。期盼對職校英語聽力學習環境、學習者、及授課教師皆有所助益。 / Video-aided English learning programs have been prevalent among students to facilitate listening skills so far. The purpose of the study is to apply a proven effective treatment of Advance Organizer (AO)—vocabulary pre-teaching and pre-questioning —with video to improve students’ listening comprehension, and to differentiate the treatment effect on students of different proficiency levels. The participants in this study are 145 first-year vocational high school students in the Departments of Chemical Engineering, Mechanical Engineering and Interior Designing. Two equivalent tests of the listening comprehension section of GEPT-Elementary Level are used as the pretest and posttest. Students are divided into high, middle and low proficiency groups according to the pretest scores. During the 16-week treatment duration, the experimental group receives AO instruction along with the supplementary videos—Family Album, U. S. A . and A+English, while the control group receives the same videos without any treatment instruction. Two questionnaires, the English Learning Orientation Survey and the Self-Reflection Questionnaire, serve to elicit background profiles of the participants to cross-examine the relations with their test performance. The results show that students receiving the AO with video instruction make significant progress on listening comprehension performance. It is also found that the effect is especially appreciable for students of high proficiency level. Based on the findings, this study proposes some pedagogical implications for industrial vocational high school teachers and indicates a direction to steer English listening instruction in vocational high schools.

No description available.

Basé sur la théorie de cycle endogène, cette thèse de doctorat étudie la question d'instabilité macro-économique et des fluctuations endogènes dans trois économies distinctes : I) avec des effets externes en consommation dans un modèle de Ramsey; II) avec des effets externes en loisir dans un modèle de générations imbriquées; et III) avec imperfection de marché du crédit dans un modèle de croissance endogène monétaire. / Based on the endogenous cycle theory, this doctoral thesis studies the issue of macroeconomic instability and fluctuations in three distinct economies: i) with consumption externalities in Ramsey model; ii) with leisure externalities in an overlapping generations model; and iii) with credit market imperfection in a monetary endogenous growth model.