教学课程资源库 更多>>

关键词： Feature extraction   Roads   Convolution   Semantics   Benchmark testing   Computer science   Convolutional neural networks   Lane marking detection   structure-aware   deep neural network   intelligent transportation system  

摘要： Lane marking detection is a fundamental task, which serves as an important prerequisite for automatic driving or driver-assistance systems. However, the complex and uncontrollable driving road environment as well as the discontinuous lane marking appearance make this task challenging. In this work, a novel deep neural network architecture is presented to detect lane markings in a complex environment by analyzing their structure information. There are two contributions to the network design. Firstly, a semantic-guided channel attention (SGCA) module is developed to select the low-level features of a deep convolutional neural network by taking the high-level features as the guidance. Secondly, a pyramid deformable convolution (PDC) module is formulated to enlarge the receptive fields and to capture the complex structures of lane markings by applying deformable convolutions on multiple feature maps with different scales. Hence, our network can better reduce false detection and enhance lane marking structures simultaneously. The experimental results on three benchmark datasets for lane marking detection show that our method outperforms other methods on all the benchmark datasets.

关键词： Security   Cyber-physical systems   Cloud computing   Data privacy   Computer science   Anomaly detection   Monitoring  

摘要： N/A

关键词： black middle-class   cultural capital   Cultural consumption   elites   strategic assimilation  

摘要： This article provides a case study of race and big-gift cultural patronage, a theoretically and empirically understudied phenomenon, by investigating million-dollar donations to the Smithsonian Institution by black patrons. I find that large donations by black supporters are concentrated at one Smithsonian museum – the National Museum of African American History and Culture. To explain this distinctive pattern of cultural consumption, I draw on ethnographic data and archival texts related to patronage at African American museums. Gifts to the National Museum of African American History and Culture can be partly explained by strategic acculturation, or an impulse to articulate and nurture black identity through consuming black culture. However, cultural steering also played a key role. Black donors were identified and cultivated via a robust fundraising infrastructure where market research cast them as key constituents of the National Museum of African American History and Culture and taste-making strategies that constructed the museum as worthy reinforced their attraction to the museum. By conceptually and empirically elaborating cultural steering, this analysis offers a more complete model of black middle- and upper-class consumption of black culture. © The Author(s) 2019.

关键词： Students  

摘要： Proven practices to recruit domestic computer science graduate students.

关键词： Computer Science   general   Computer Hardware   Computer Systems Organization and Communication Networks   Software Engineering/Programming and Operating Systems   Data Structures and Information Theory  

摘要： Zusammenfassung Technologie und digitale Geschäftsmodellen sind zentrale Grundlagen für nachhaltige Lebensqualität und Wettbewerbsfähigkeit. Akzeptanz, Adaption und überzeugende Narrative sind Voraussetzungen, um digitale Wertschöpfung zu realisieren.

关键词： Frequency modulation   Indexes   Machine learning   Tools   Computer science   Linguistics   XAI   explainable artificial intelligence   data fusion   information fusion   deep learning  

摘要： The modern era of machine learning is focused on data-driven solutions. While this has resulted in astonishing leaps in numerous applications, explainability has not witnessed the same growth. The reality is, most machine learning solutions are black boxes. Herein, we focus on data/information fusion in machine learning. Specifically, we explore four eXplainable Artificial Intelligence (XAI) questions relative to Choquet integral;(i) what is the quality of our inputs and their interactions, (ii) how is the information being combined, (iii) what is the quality of our training data (and thus our learned models), and (iv) what trust do we place in an output? Previously, we derived an initial set of indices for (i)-(iv) on the premise of perfect knowledge. Herein, we make XAI more accurate by taking into consideration what the machine learned. A combination of synthetic data and real-world experiments from remote sensing for fusing deep learners in the context of classification are explored. Our approach leads to performance gain, insights into what was learned, and it helps us realize better future solutions.

关键词： Education   Systematics   Proposals   Data mining   Terminology   Teamwork   Planning   Active learning   computer science   problem-based learning (PBL)   systematic mapping  

摘要： Contribution: This article adds to the results of previous systematic mapping study by addressing a more ample context of problem-based learning (PBL) in computing education. Background: PBL is defined as an instructional method of constructivist teaching that uses real problems as a motivating element for learning. Although PBL was born in medical education, it has been used in computing education to facilitate the students' engagement and learning capacity, contributing to developing skills, such as teamwork, holistic vision, critical thinking, and solving problem. Considering that approach much more descriptive than prescriptive, it favors the implementation of diverse methodologies on its behalf. Research Question: "How the PBL approach has been applied in teaching computing?" Methodology: From the initial set of 3837 studies, 102 primary studies were selected and categorized through Kitchenham methodology for analysis from 1999 to 2019. Findings: Based on these studies, it was possible to verify a greater adoption of PBL in undergraduate education and a great diversity of strategies involving real-world problems, developing technical and no-technical skills and making the student much more motivated and engaged. Challenges were also mapped, which shows a very interesting field for new research.

关键词： Semantics   Training   Manifolds   Training data   Bridges   Computer science   Correlation   Cross-modal hashing   imbalanced multi-modal data   adversarial tri-fusion hashing   manifold structure  

摘要： Cross-modal retrieval has received increasing attentions for efficient retrieval across different modalities, and hashing technique has made significant progress recently due to its low storage cost and high query speed. However, most existing cross-modal hashing works still face the challenges of narrowing down the semantic gap between different modalities and training with imbalanced multi-modal data. This article presents an efficient Adversarial Tri-Fusion Hashing Network (ATFH-N) for cross-modal retrieval, which lies among the early attempts to incorporate adversarial learning for working with imbalanced multi-modal data. Specifically, a triple fusion network associated with zero padding operation is proposed to adapt either balanced or imbalanced multi-modal training data. At the same time, an adversarial training mechanism is leveraged to maximally bridge the semantic gap of the common representations between balanced and imbalanced data. Further, a label prediction network is utilized to guide the feature learning process and promote hash code learning, while additionally embedding the manifold structure to preserve both inter-modal and intra-modal similarities. Through the joint exploitation of the above, the underlying semantic structure of multimedia data can be well preserved in Hamming space, which can benefit various cross-modal retrieval tasks. Extensive experiments on three benchmark datasets show that the proposed ATFH-N method yields the comparable performance in balanced scenario and brings substantial improvements over the state-of-the-art methods in imbalanced scenarios.

关键词： Education   Engineering profession   Psychology   Task analysis   Instruments   Computer science   Encoding   Computer science (CS)   motivation   self-regulated learning (SRL)   undergraduate  

摘要： Contribution: This article presents a synthesis of the findings and implications from the IC2Think program of research in undergraduate computer science (CS) courses examining student motivation and self-regulated learning (SRL). These studies illuminate both the difficulty and potential for motivating CS students, as well as the uniqueness of CS as a context for studying undergraduate motivation. Background: Computing disciplines are increasingly important in preparing the future workforce. It is imperative that CS educators understand how to motivate students and enhance student outcomes. Synthesizing findings across multiple studies allows for the emergence of new insights into student motivation and SRL. Research Questions: Which aspects of students' motivation and SRL are predictive of achievement and retention in CS and how can findings inform CS education? Methodology: The primary methodology is a comprehensive review of seven years of research on undergraduate CS education. Studies use a variety of analysis techniques, examine a range of constructs, and include multiple introductory and advanced CS courses. Studies of relationships between variables and change over time were conducted. Findings: The present synthesis of studies on motivation and SRL highlights the complex, counter-intuitive, and positive aspects of student motivation in CS.