教学课程资源库 更多>>

关键词： Artificial intelligence   Physics   Engineering   Computer science   Electrical engineering   Mathematics   Optics  

摘要： The ultimate goal of this research is to enable efficient, reliable, and cost-effective monitoring of plate-like wing or fuselage structures using sparse sensor networks. In ultrasonic guided wave (UGW) based structural health monitoring (SHM), physics-based and data-driven (learning-based) methods represent two ends of the spectrum for damage diagnosis. Physics-based methods rely primarily on the underlying physics in the guided waves in order to draw inferences about the structural health. These methods have witnessed a lot of research and development over the past decade, however, when applied to practical structural components under realistic environmental and operational conditions (EOCs), they are rendered highly insufficient. Data driven learningbased methods draw inferences directly from the sensed data and circumvent some of the limitations of physics-based methods, however, being physics-agnostic, they are highly data intensive. At present, with sparse sensor networks, both physics-based and learning-based methods can only achieve low-level damage assessment (damage existence and localization) in rather simple plate-like structures. This bottleneck is caused by sparse sensor networks providing limited information about the guided wave propagating within the structure, with low signal-to-noise ratio (SNR) and low spatial resolution of the measured sensor data being the two main contributing factors. In contrast, full guided wavefield images provide rich information about the guided wave propagation, including the interactions of the propagating wave with damages and/or geometric features within the structure. However, full wavefield scanning at high spatial resolution using say air-coupled transducers (ACTs) or laser Doppler vibrometers (LDVs) demands a prohibitively long scanning time. Ideally, high speed cameras can circumvent this problem, however, the sensing resolution remains low unless the field of view is small. A paradigm shift, involving the integ

关键词： Information science \u0026 technology   computer science   information science \u0026 technology   computing: general   information science \u0026 technology   computer science  

ISBN: (数字)9781636393056 ISBN: (纸本)9781636393063;9781636393049

摘要： This book introduces readers to emerging persistent memory (PM) technologies that promise the performance of dynamic random-access memory (DRAM) with the durability of traditional storage media, such as hard disks and solid-state drives (SSDs). Persistent memories (PMs), such as Intel"s Optane DC persistent memories, are commercially available today. Unlike traditional storage devices, PMs can be accessed over a byte-addressable load-store interface with access latency that is comparable to DRAM. Unfortunately, existing hardware and software systems are ill-equipped to fully avail the potential of these byte-addressable memory technologies as they have been designed to access traditional storage media over a block-based interface. Several mechanisms have been explored in the research literature over the past decade to design hardware and software systems that provide high-performance access to *** PMs are durable, they can retain data across failures, such as power failures and program crashes. Upon a failure, recovery mechanisms may inspect PM data, reconstruct state and resume program execution. Correct recovery of data requires that operations to the PM are properly ordered during normal program execution. Memory persistency models define the order in which memory operations are performed at the PM. Much like memory consistency models, memory persistency models may be relaxed to improve application performance. Several proposals have emerged recently to design memory persistency models for hardware and software systems and for high-level programming languages. These proposals differ in several key aspects; they relax PM ordering constraints, introduce varying programmability burden, and introduce differing granularity of failure atomicity for PM *** primer provides a detailed overview of the various classes of the memory persistency models, their implementations in hardware, programming languages and software systems proposed in the recent res

关键词： Biomedical engineering   Computer engineering   Computer science   Mental health  

摘要： Technological solutions to screen for depression are being researched to combat the rise of mental health problems in post-secondary students. Specifically, machine learning models can be trained on smartphone datasets to detect depression. However, model performance is hindered by limitations of existing datasets, such as: (1) sparse depression labels and (2) neglecting individual circumstances variables. To address the problem with sparse labels, we explored multiple instance learning on an existing sparsely labelled dataset. We also collected a new dataset containing more frequent depression labels and individual circumstances variables using a custom-built mobile application. Overall, our best performing multiple instance learning model outperformed other models with 0.661 AUC. We also achieved up to 0.992 AUC in depression detection using our own dataset. Ultimately, this work contributed to building a better quality smartphone dataset that can train more accurate depression detection machine learning models, helping future students seek timely care.

关键词： Artificial intelligence   Computer science   Computer vision   Deep learning   Medical image analysis   Representation learning  

摘要： Deep Learning approaches have achieved revolutionary performance improvement on many computer vision tasks from understanding natural images and videos to analyzing medical images. Besides building more complex deep neural networks (DNNs) and collecting giant annotated datasets to obtain performance gains, more attention has now been focused on the shortcomings of DNNs. As recent research has shown, even when trained on millions of labeled samples, deep neural networks may still lack robustness to domain shift, small perturbations, and adversarial examples. On the other hand, in many real-world scenarios, e.g. in clinical applications, the number of labeled training samples is significantly smaller than for large existing deep learning benchmarks. Moreover, current deep learning models cannot generalize to samples with novel combinations of seen elementary concepts. Therefore, in this thesis, I focus on handling the critical needs to make modern deep learning approaches applicable in the real-world with a focus on computer vision tasks. Specifically, I focus on data efficiency, robustness, and generalization. I propose (1) DeepAtlas, a joint learning framework for image registration and segmentation that can learn DNNs for both tasks from unlabeled images and a few labeled images. (2) RandConv, a data augmentation technique that applies a random convolution layer on images during training to improve the generalization performance of a DNN in the presence of domain shift and robustness to image corruptions. (3) CompGen, a comprehensive study of compositional generalization in unsupervised representation learning on disentanglement and emergent language models.

关键词： Cloud computing   Fault tolerant systems   Fault tolerance   Load management   Systematics   Computer science   Load modeling   Dynamic load balancing   reactive fault tolerance   cloud   systematic literature review  

摘要： The computational cloud aims to move traditional computing from personal computers to cloud providers on the internet. Cloud security represents an important research area. Confidentiality, integrity, and availability are the main cloud security characteristics addressed. Cloud providers apply dynamic load balancing and reactive fault tolerance techniques to build secure cloud services to achieve high service availability. Dynamic cloud load balancing approaches distribute submitted tasks to virtual machines during tasks execution and the load of virtual machines is updated based on the system's state. Reactive cloud fault tolerance is activated for system process failures after failure effectively happens. Reactive cloud fault tolerance handles failure after the fault has occurred. Despite the significance of dynamic load balancing and reactive fault tolerance techniques and mechanisms, few reviews focus on these approaches in a systematic, unbiased method focusing on integrating cloud dynamic load balancing and reactive fault tolerance techniques. This paper conducts a systematic literature review of the existing literature concerning reactive fault tolerance, dynamic load balancing, and their integration in their basic approaches, types, frameworks, and future directions.

关键词： American culture   Black culture   Black experience   Collage   Sculpture  

摘要： This text is one explanation of how the Black experience and culture has shaped American history. Utilizing illustrated conceptual and structural artistic frameworks of Afro-Surrealism and Surrealism pushes the message along. By assessing the influence of Black sports figures, creative thinkers, civil movements, then interweaving personal narratives a clear understanding of how American history evolved due to the advancement and resiliency of Black people. Looking at artists like Romare Bearden and Arthur Jafa their creative interpretation of the Black community’s existence emphasized the strength of difference within the culture itself. Sports figures like Colin Kaepernick and Muhammad Ali help to depict how the role of celebrity challenges social and civil unrest. While my own personal narrative brings in a view of the advantages and opportunities afforded in todays present era. Revealing how the influence of Black culture continues and has been a shaping force for America’s global presence.

关键词： User experience   Augmented reality   Indexes   Computer science   User experience   Systematics   User interfaces   Augmented reality (AR)   category   quality attributes   user experience (UX)   user interface (UI)  

摘要： This study proposes a characterization of quality attributes for applications that use augmented reality. This classification is done from the perspective of the user experience. The attribute identification was based on primary studies of the IEEE Xplore, Scopus, and ACM repositories. From an initial set of 1165 papers, 101 documents were selected. The document proposes two categories: objective and subjective. In the objective category 4 subcategories and 40 attributes were found, and in the subjective one 5 subcategories and 54 attributes were found. This is the first time that all these criteria are presented in a single document, which is the input for designing a comprehensive quality assurance tool for user experience in augmented reality.

关键词： Physiology   Analytical chemistry   Bioinformatics   Biology   Chemistry   Computer science   Genetics   Geochemistry   Mineralogy  

摘要： Scleractinian corals are evolutionary-successful calcifying marine organisms, which utilize an endo-symbiotic relationship with photosynthetic dinoflagellate algae that transfer their photosynthates products to their coral hosts. Coral biomineralization is one of the fundamental processes in the ocean, in which the coral polyps build approximately 1012 kg per year of calcium carbonate structures. Previous studies have been suggested that the mineral deposition is enhanced by the light (also known as the 'light enhanced calcification' phenomena) due to the high transportation of energy products of the dinoflagellate endosymbiont. In addition, studies of skeletal formation have shown two types of mineral depositions, which are referred to be formed across the diel cycle. During the nighttime, the mineral is extended with nano-granules particles to develop a framework for the daytime precipitation of needle shape fibers that are rapidly growing and filling the space between the extended gaps. However, some studies present this process as controversial. Here, I show the biological involvement in the mineral formation across the diel cycle. My results indicate that Light enhanced calcification is related to several processes that occur across the diel cycle; during nighttime, tissue might elevate away from the skeleton, extending the calcifying space area to enable the formation of a new organic framework template. During the daytime, the combination of synthesis of acid-rich proteins and a greater flux of calcium and carbonate ions to the sites of calcification facilitate the conditions for extensive mineral growth. Corals have a biphasic life cycle with planktonic larval stages and benthic adults' phases. These two phases are separated by settlement and metamorphosis, which are critical stages in coral development. The planulae undergo intense changes in morphology, the development of new tissues, and, in some cases, the uptake of symbionts. Initial mineral deposits we

关键词： Mathematics   Computer science  

关键词： Optimization   Neural networks   Support vector machines   Data processing   Geophysics   Earthquakes   Optics   Artificial intelligence   Computer science   Plate tectonics  

摘要： As new seismic acquisition methods arise, growing data volumes call for automated processing methods to extract full value out of the recorded data. Herein, we develop an end-to-end machine learning framework for seismic event detection and identification on continuous data. We illustrate our methodology through three field-data use cases. Firstly, we perform earthquake detection using fiber-optic cables in the telecommunication conduits under the Stanford University campus. We identify new uncataloged small-magnitude local earthquakes by analyzing more than three years of continuous recordings. We demonstrate that fiber-optic cables can complement sparse seismometer networks. We then tackle microseismic event detection in fiber-optic data acquired inside an unconventional reservoir. Our methodology identifies more than 100,000 events over ten hydraulic stimulation stages, allowing the reconstruction of the spatio-temporal fracture development far more accurately and efficiently than would have been feasible by traditional methods. Finally, we explore tectonic tremor identification using a catalog of more than 1 million events detected along the central San Andreas Fault over a period of 15 years. Tectonic tremors are composed of hundreds of repeating low-frequency earthquakes (LFEs). These LFEs are near the noise level and are thus usually found via a multichannel matched-filter search using carefully curated waveform templates. We demonstrate that our methodology can successfully detect new LFEs with low signal amplitude without a prior template.