In the past decade software products have become pervasive in many aspects of people's lives around the world. Unfortunately, the quality of the experience an individual has interacting with that software is depen...
详细信息
ISBN:
(纸本)9781450385688
In the past decade software products have become pervasive in many aspects of people's lives around the world. Unfortunately, the quality of the experience an individual has interacting with that software is dependent on the quality of the software itself, and it is becoming more and more evident that many large software products contain a range of issues and errors, and these issues are not known to the developers of these systems, and they are unaware of the deleterious impacts of those issues on the individuals who use these systems. The authors of this paper are developing a new digital ethics curriculum for the instruction of computer science students. In this paper we present case studies that were explored to demonstrate programming issues to First Year Computer Science students. Each case study outlines key issues associated with a particular scenario and is accompanied by specific questions to be used by the instructor to allow students to begin to reflect on, and evaluate, the implications of these issues. The objective of this teaching content is to ensure that the students are presented with, and engage with, ethical considerations early in their studies and well before they encounter them in an employment setting.
Machine Learning (ML) is increasingly being adopted in different industries. Deep Reinforcement Learning (DRL) is a subdomain of ML used to produce intelligent agents. Despite recent developments in DRL technology, th...
详细信息
Machine Learning (ML) is increasingly being adopted in different industries. Deep Reinforcement Learning (DRL) is a subdomain of ML used to produce intelligent agents. Despite recent developments in DRL technology, the main challenges that developers face in the development of DRL applications are still unknown. To fill this gap, in this paper, we conduct a large-scale empirical study of 927 DRL-related posts extracted from Stack Overflow, the most popular Q &A platform in the software community. Through the process of labeling and categorizing extracted posts, we created a taxonomy of common challenges encountered in the development of DRL applications, along with their corresponding popularity levels. This taxonomy has been validated through a survey involving 65 DRL developers. Results show that at least 45%\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$45\%$$\end{document} of developers experienced 18 of the 21 challenges identified in the taxonomy. The most frequent source of difficulty during the development of DRL applications are Comprehension, API usage, and Design problems, while Parallel processing, and DRL libraries/frameworks are classified as the most difficult challenges to address, with respect to the time required to receive an accepted answer. We hope that the research community will leverage this taxonomy to develop efficient strategies to address the identified challenges and improve the quality of DRL applications
Recent advances in deep learning promote the innovation of many intelligent systems and applications such as autonomous driving and image recognition. Despite enormous efforts and investments in this field, a fundamen...
详细信息
ISBN:
(纸本)9781728149820
Recent advances in deep learning promote the innovation of many intelligent systems and applications such as autonomous driving and image recognition. Despite enormous efforts and investments in this field, a fundamental question remains under-investigated-what challenges do developers commonly face when building deep learning applications? To seek an answer, this paper presents a large-scale empirical study of deep learning questions in a popular Q&A website, Stack Overflow. We manually inspect a sample of 715 questions and identify seven kinds of frequently asked questions. We further build a classification model to quantify the distribution of different kinds of deep learning questions in the entire set of 39,628 deep learning questions. We find that program crashes, model migration, and implementation questions are the top three most frequently asked questions. After carefully examining accepted answers of these questions, we summarize five main root causes that may deserve attention from the research community, including API misuse, incorrect hyperparameter selection, GPU computation, static graph computation, and limited debugging and profiling support. Our results highlight the need for new techniques such as cross-framework differential testing to improve software development productivity and software reliability in deep learning.
This paper reviews some of the author's experiences during the past 20 years of treating severe problem behavior. Factors that represent barriers to success are identified and discussed, as are factors that contri...
详细信息
This paper reviews some of the author's experiences during the past 20 years of treating severe problem behavior. Factors that represent barriers to success are identified and discussed, as are factors that contribute to the development of successful treatment programs. Barriers to success include the inherent reactive nature of human services and educational systems, expertise problems, systems problems, information gaps, programming problems, characteristics of problem behavior, and maintenance problems. Some new programmatic directions are suggested for overcoming the various treatment barriers. The paper concludes with strategies and factors to consider that will ensure long-term success in the treatment of severe problem behavior.
Purpose Improving service quality, student satisfaction and student loyalty is important to higher education institutions’ sustainable growth. The objectives of this study are a twofold: first, the study seeks to de...
详细信息
Purpose Improving service quality, student satisfaction and student loyalty is important to higher education institutions’ sustainable growth. The objectives of this study are a twofold: first, the study seeks to determine the dimensions of higher education service quality with a specific focus on Vietnam. Second, it examines how the service quality dimensions impact student satisfaction and student loyalty, with the moderating role of the university image. Design/methodology/approach This study followed a rigorous procedure, including interviews, a survey, exploratory factor analysis (EFA) and reliability analysis to identify higher education service quality dimensions and their measures. After that, using the data obtained from 1,550 university students in Vietnam, confirmatory factor analysis was used to validate the identified dimensions and structural equation modeling was used to test a proposed model explaining the outcomes of higher education service quality. Findings The findings reveal five dimensions of higher education service quality: academic aspect, nonacademic aspect, programming issues, facilities and industry interaction. Most of these factors have a positive influence on student satisfaction. In addition, the university image moderates the positive relationship between student satisfaction and student loyalty. Practical implications This study’s findings highlight the complexity of service quality in the higher education context and encourage higher education institutions to improve their service quality in image to enhance student satisfaction and loyalty. Originality/value This study suggests a unique measure of higher education service quality dimensions and provides fresh insights into how they impact student satisfaction and loyalty in Vietnam.
Constructive alignment has been widely accepted as a strong pedagogical approach that promotes deep learning, however its application to programming units in higher education has not been widely reported. A constructi...
详细信息
ISBN:
(纸本)9781921770210
Constructive alignment has been widely accepted as a strong pedagogical approach that promotes deep learning, however its application to programming units in higher education has not been widely reported. A constructively aligned introductory programming unit with portfolio assessment provides an opportunity for students to reflect on their learning. These reflections provide a rich source of information for educators looking to identify topical and pedagogical issues influencing student outcomes. In this work we applied thematic analysis to the reflective reports presented by students as part of their portfolio submission for an introductory programming unit. The analysis indicates several interesting aspects related to both topical and pedagogical issues. These results can be used to inform the development of constructively aligned programming units, and inform future research.
暂无评论