教学课程资源库 更多>>

关键词： Artificial intelligence   Computer science  

摘要： The study of non-verbal cues is an important step towards human behavioral understanding. Facial expressions are key indicators of the internal state of mind. The Facial Action Coding System (FACS) is a commonly used standard to systematically annotate expressions and breaks down the prototypical expressions into individual facial muscle activations. FACS is a comprehensive system that encodes all possible facial muscle movements using primitives known as Action Units (AUs).FACS is extensively used in the development of computer vision-based affective computing applications. Automated recognition of Action Units (AU) is useful for applications in diverse areas such as in medicine for measurement of pain and analysis of depression, and in law enforcement for deceit detection, among other fields. Automated AU recognition is still an unsolved problem in the computer vision domain due to several impeding factors. We identify three critical challenges in AU detection for deep learning-based frameworks: 1. Action Units (AUs) are very subtle and local, making them hard to annotate even for trained human coders. 2. An expression is typically seen as a combination of AUs. Several AUs can co-occur at a given instant. Discerning the associations between individual AUs is a complex task requiring an understanding of facial anatomy. 3. Although expressions cause discernible appearance changes on the face, tracking the minuscule muscle movements can be very challenging in the presence of factors such as illumination changes, head pose variations, and occlusions. Additional sources such as thermal images and physiological signals can provide auxiliary semantic information that could prove beneficial. However, their usefulness has never been explored in AU recognition. The goal of this dissertation is to address these challenges using a three-pronged approach: Attention Learning, Relation Modelling, and Multimodal Learning. Humans have the innate ability to rapidly spot and react t

关键词： Information science \u0026 technology   computer science  

ISBN: (数字)9781636392691 ISBN: (纸本)9781636392707;9781636392684

摘要： This book introduces Participatory Design to researchers and students in Human–Computer Interaction (HCI). Grounded in four strong commitments, the book discusses why and how Participatory Design is important today. The book aims to provide readers with a practical resource, introducing them to the central practices of Participatory Design research as well as to key references. This is done from the perspective of Scandinavian Participatory Design. The book is meant for students, researchers, and practitioners who are interested in Participatory Design for research studies, assignments in HCI classes, or as part of an industry project. It is structured around 11 questions arranged in 3 main parts that provide the knowledge needed to get started with practicing Participatory Design. Each chapter responds to a question about defining, conducting, or the results of carrying out Participatory Design. The authors share their extensive experience of Participatory Design processes and thinking by combining historical accounts, cases, how-to process descriptions, and reading lists to guide further readings so as to grasp the many nuances of Participatory Design as it is practiced across sectors, countries, and industries.

摘要： The gender imbalance in computer science is getting worse, with computer science being the only Science, Technology, Engineering, and Mathematics (STEM) field in which women’s representation has steadily declined throughout the past few decades. Rectifying this gender imbalance is an urgent need, both because of fundamental issues of equity and because leaving most women out of the labor market in computer-related industries dramatically lowers the number of American workers to fill these roles. Although many efforts have proved to be effective in increasing women’s participation in computer science, many of those may be difficult to replicate on a large scale, as they relied on a great deal of investment of time and money from high-level administrators. Therefore, it is worthwhile to consider alternatives. My dissertation focuses on one attempt to redress the lack of women in computer science—Women-in-Computer-Science (Women-in-CS) or Women-in-Computing (WiC) college clubs. Drawing on organizational change theories, I ask: What are some characteristics of the Women-in-CS clubs and how do students engage them? Do they engender changes? If so, how? I explored these research questions through two theoretically connected but methodologically different studies, using thematic network analysis and ethnography respectively. Through a cross-level analysis, I argue that Women-in-CS clubs’ mission statements and practices directly speak to the challenges of gender inequality in computer science. For students, Women-in-CS club is an identity, a community, a safe space, and the beginning of a career-long practice of mentoring future women in computer science. Moreover, changes driven by such clubs have brought impacts to both the individual level female computer science students and the organizational level computer science departments. Both top-down and bottom-up leadership approaches have been used by these clubs to overcome resistance and obstacles related to social cogniti

关键词： Medical imaging   Artificial intelligence   Computer science   Biomedical engineering  

摘要： Cardiovascular disease is the leading cause of death in US and non-invasive cardiac imaging has vital importance for early detection and diagnosis of heart disease. Cardiac Magnetic Resonance (CMR) is arguably the most versatile imaging modality and capable of a comprehensive evaluation of heart disease without ionization radiation. Despite the advantages of CMR, it is seldom used (only 1% footprint) due to lack of availability and higher cost, which is mainly caused by the long scan times. Meanwhile, many accelerated CMR acquisition and quantification methods were feasible, but their clinical translation was limited due to long image reconstruction time and long manual processing time (e.g. segmentation). The purpose of this dissertation was to describe the development and validation of accelerated CMR methods using compressed sensing (CS) and deep learning (DL) to overcome current limitations. This dissertation includes the following topics: (i) CS image reconstruction of accelerated coronary quiescent-interval slice-selective (QISS) magnetic resonance angiography (MRA) that enabled single-shot, free-breathing acquisition, (ii) a newly developed high-resolution late gadolinium enhancement (LGE) CMR sequence with novel CS image reconstruction that provides multi-TI image contrast, (iii) rapid image reconstruction of highly undersampled real‐time cine using deep learning without significant loss in image quality, visual scores and functional parameters, (iv) automated image segmentation of biventricular tissue phase mapping (TPM) using deep learning.

关键词： Computer science  

摘要： Social media has become an integral part of human life in the 21st century. The number of social media users was estimated to be around 3.6 billion individuals in 2020. Social media platforms (e.g., Facebook) have facilitated interpersonal communication, diffusion of information, the creation of groups and communities, to name a few. As far as education systems are concerned, online social media has transformed and connected traditional social networks within the schoolhouse to a broader and expanded world outside. In such an expanded virtual space, teachers engage in various activities within their communities, e.g., exchanging instructional resources, seeking new teaching methods, engaging in online discussions. Therefore, given the importance of teachers in social media and its tremendous impact on PK-12 education, in this dissertation, we investigate teachers in social media from a data science perspective. Our investigation in this direction is essentially an interdisciplinary endeavor bridging modern data science and education. In particular, we have made three contributions, as briefly discussed in the following. Current teachers in social media studies suffice to a small number of surveyed teachers while thousands of other teachers are on social media. This hinders us from conducting large-scale data-driven studies pertinent to teachers in social media. Aiming to overcome this challenge and further facilitate data-driven studies related to teachers in social media, we propose a novel method that automatically identifies teachers on Pinterest, an image-based social media popular among teachers. In this framework, we formulate the teacher identification problem as a positive unlabelled (PU) learning where positive samples are surveyed teachers, and unlabelled samples are their online friends. Using our framework, we build the largest dataset of teachers on Pinterest. With this dataset at our disposal, we perform an exploratory analysis of teachers on Pinterest

关键词： Mathematics   Theoretical mathematics   Computer science  

摘要： We solve the problem of finding Hamiltonian paths on rectangular grids that have the minimum possible number of turns. It is found that the minimum number of turns possible for a Hamiltonian path on an m x n rectangular grid is 2M—2, where M is the minimum of m and n. A method for enumerating minimum turn Hamiltonian paths is presented and a computer algorithm is used to count how many minimum turn Hamiltonian paths exist for M = 2,3,...,13. Additionally, a computer algorithm is used to list some minimum turn Hamiltonian paths. Lastly, a connection to stamp folding and meanders is found.

关键词： Quantum physics   Computer science   Physics   Computational physics  

摘要： Quantum computing is the offspring of quantum mechanics and computer science, two great scientific fields founded in the 20th century. Quantum computing is a relatively young field and is recognized as having the potential to revolutionize science and technology in the coming century. The primary question in this field is essentially to ask which problems are feasible with potential quantum computers and which are not. In this dissertation, we study this question with a physical bent of mind. We apply tools from computer science and mathematical physics to study the complexity of simulating quantum systems. In general, our goal is to identify parameter regimes under which simulating quantum systems is easy (efficiently solvable) or hard (not efficiently solvable). This study leads to an understanding of the features that make certain problems easy or hard to solve. We also get physical insight into the behavior of the system being simulated. In the first part of this dissertation, we study the classical complexity of simulating quantum dynamics. In general, the systems we study transition from being easy to simulate at short times to being harder to simulate at later times. We argue that the transition timescale is a useful measure for various Hamiltonians and is indicative of the physics behind the change in complexity. We illustrate this idea for a specific bosonic system, obtaining a complexity phase diagram that delineates the system into easy or hard for simulation. We also prove that the phase diagram is robust, supporting our statement that the phase diagram is indicative of the underlying physics. In the next part, we study open quantum systems from the point of view of their potential to encode hard computational problems. We study a class of fermionic Hamiltonians subject to Markovian noise described by Lindblad jump operators and illustrate how, sometimes, certain Lindblad operators can induce computational complexity into the problem. Specifically, we sh

关键词： Coronaviruses   Bone cancer   Neural networks   COVID-19   Artificial intelligence   Computer science  

关键词： Computer science   Artificial intelligence  

摘要： In this thesis, I will discuss the problem of acquiring visual intelligence from the interaction, focusing on two aspects of visual understanding: (1) visual perception and (2) embodied intelligence. To address the first question, I designed experiments to learn visual representations by observing animals and humans interact with the visual world. Further, I investigated the idea of learning perception from hands-on interaction -- acquiring generalizable physical understanding by predicting the forces applied in an observed video and trying to replicate the motion observed in simulation, with no additional supervision provided. To address the second question, I discuss our findings on training intelligent embodied agents using interaction from two perspectives. I designed a training paradigm that enables learning-to-learn from interactions. This training regime helps us to continue to learn from our interactions even during inference time. Moreover, I introduce a visually rich object manipulation framework, ManipulaTHOR, which opens the gate for directly training embodied agents to interact intelligently in a physically realistic environment via low-level object manipulation and navigation.

关键词： Adaptive support   Collaborative learning   Pair programming   Student-student dialogue   Uncertainty   Wizard-of-Oz  

摘要： In computer science education, introductory computer programming courses tend to be the most challenging for novices, with higher dropout rates than other computer science courses. Recruitment and retention of students in computer science fields is an important area of focus in computer science education research, and previous research has considered many classroom approaches. Collaborative practices in the classroom in particular have contributed to lower dropout rates, with structured practices such as pair programming resulting in higher quality code products and increasing retention of students in computer science majors. While pair programming has shown many benefits, there are still challenges to consider. Incompatible partners and getting stuck during collaboration both contribute negatively to the learning experience. Instructors can help mitigate some situations, but this solution is not scalable to large classrooms. Adaptive support systems specifically designed to support collaboration show great promise, achieving similar outcomes to human tutors. The collaborative process must be investigated more deeply in order to determine how exactly such systems can best support students. The goal of this dissertation work was to advance the state of knowledge around how undergraduate students collaborate on computer science tasks through pair programming. In my work, I have modeled and compared the problem-solving approaches of individual students and students pairs, examined the differences between the dialogue of more successful and less successful pairs, and revealed the ways in which learners expressed and addressed uncertainty. My work has identified three recommendations for how to best support students during pair programming in the context of addressing and resolving uncertainty: students should 1) notify partner when uncertainty is resolved, 2) focus on resolving current uncertainty before proceeding with the task, and 3) explain their current thought pro