教学课程资源库 更多>>

关键词： Artificial intelligence   Computer science   Gerontology   Mechanics   Medicine   Morphology   Neurosciences   Physics  

摘要： Balance deficits can lead to injurious falls, early enrollment in long-term care facilities and hospitalization. Multiple clinical tools have been developed to identify balance deficits and direct treatment recommendations; however, they are often based on simple measures which might fail to sufficiently characterize an individual’s balance ability. Force plate systems can be used to evaluate balance in a laboratory setting; however, their non-portability renders them difficult for use in the clinic and in the field. Markerless motion capture is an emerging technology which could bridge the gap between clinical and laboratory assessments of balance by offering clinicians a relatively low-cost and versatile method of obtaining high resolution measurements of balance ability. This work aims to validate the use of a deep neural network-, video-based markerless motion capture system, Theia3D, to measure aspects of balance which are commonly identified using force plate systems, as well as aspects of postural control which cannot be identified using standard clinical or laboratory tools. Three studies were carried out as part of this thesis: (1) an algorithm was developed to automatically identify distinct balance tasks within a functional assessment circuit, (2) measures of balance performance quantified using Theia3D were directly and indirectly compared with measures from a force plate system, and (3) the relationships between whole-body kinematics and participant age, BMI, and sex and measures from a force plate system during the maintenance of static balance poses were investigated. Results from these studies demonstrate that Theia3D can be used (1) to automatically identify performances of distinct functional tasks, (2) to measure aspects of balance which are commonly identified using force plate systems, with similar levels of reliability, and (3) to measure whole-body motion to uncover age-, BMI-, and sex-specific mechanisms of balance which would otherwise be un

关键词： Computer science   Random access memory   Accuracy   Deep learning   Application programming interface   Datasets   Artificial intelligence   Neural networks   Design techniques   Bottlenecks  

摘要： Convolutional Neural Network (CNN) models have become the mainstream method in Artificial Intelligence (AI) areas for computer vision tasks like image classification and image segmentation. Deep CNNs contain a large volume of convolution calculations. Thus, training a CNN requires powerful GPU resources. Training a large CNN may take days or even weeks, which is time-consuming and costly. When we need multiple runs to search for the optimal CNN hypermeter settings, it would take a couple of months with limited GPUs, which is not acceptable and hinders the development of CNNs. It is essential to train CNN faster. There are two kinds of methods to train CNN faster when no additional computing resources are available. The first method is to do the model compression, either by reducing parameters or using less storage to represent the models. This method reduces training time by reducing the architecture complexity. The second method is to reduce the input data feed into the network without affecting the network architecture. Architecture complexity reduction is a popular research area to train CNN faster. Nowadays, mobile devices like smartphones and smart cars rely on deep CNNs to accomplish complex tasks like human body recognition and face recognition. Due to the high real-time demands and the memory constraints for mobile device applications, conventional large CNN is not suitable. CNN model compression is a trend to train a deep CNN model with less computation cost. Currently, there are many successful networks designed to solve this problem, like ResNeXt, MobileNet, ShuffleNet, and GhostNet. They use 1×1 convolution, depthwise convolution, or group convolution to replace the standard convolution to reduce the computation. However, there are fewer studies on the following questions. First, does the variety of convolution layers (the output channel number is larger or smaller than the input channel number) affect different compression strategies’ performance? Secon

关键词： Propagation   Active learning   Deep learning   Algorithms   Ablation   Artificial intelligence   Computer science   Physics  

摘要： Entity Resolution refers to the process of identifying records which represent the same real-world entity from one or more datasets. In the big data era, large numbers of entities need to be resolved, which leads to several key challenges, especially for learning-based ER approaches: (1) With the number of records increasing, the computational complexity of the algorithm grows exponentially. (2) Quite a number of samples are necessary for training, but only a limited number of labels are available, especially when the training samples are highly imbalanced. Blocking technique helps to improve the time efficiency by grouping potentially matched records into the same block. Thus to address the above two challenges, in this thesis, we first introduce a novel blocking scheme learning approach based on active learning techniques. With a limited label budget, our approach can learn a blocking scheme to generate high quality blocks. Two strategies called active sampling and active branching are proposed to select samples and generate blocking schemes efficiently. Additionally, a skyblocking framework is proposed as an extension, which aims to learn scheme skylines. In this framework, each blocking scheme is mapped as a point to a multi-dimensional scheme space where each block-ing measure represents one dimension. A scheme skyline contains blocking schemes that are not dominated by any other blocking schemes in the scheme space. We develop three scheme skyline learning algorithms for efficiently learning scheme skylines under a given number of blocking measures and within a label budget limit. While blocks are well established, we further develop the Learning To Sample approach to deal with the second challenge, i.e. training a learning-based active learning model with as mall number of labeled samples. This approach has two key components: a sampling model and a boosting model, which can mutually learn from each other in iterations to improve the performance of each other

关键词： Artificial intelligence   Computer science   Computer vision   Machine learning   Neural networks   Robustness  

摘要： We introduce several new datasets namely ImageNet-A/O and ImageNet-R as well as a synthetic environment and testing suite we called CAOS. ImageNet-A/O allow researchers to focus in on the blind spots remaining in ImageNet. ImageNet-R was specifically created with the intention of tracking robust representation as the representations are no longer simply natural but include artistic, and other renditions. The CAOS suite is built off of CARLA simulator which allows for the inclusion of anomalous objects and can create reproducible synthetic environment and scenes for testing robustness. All of the datasets were created for testing robustness and measuring progress in robustness. The datasets have been used in various other works to measure their own progress in robustness and allowing for tangential progress that does not focus exclusively on natural accuracy. Given these datasets, we created several novel methods that aim to advance robustness research. We build off of simple baselines in the form of Maximum Logit, and Typicality Score as well as create a novel data augmentation method in the form of DeepAugment that improves on the aforementioned benchmarks. Maximum Logit considers the logit values instead of the values after the softmax operation, while a small change produces noticeable improvements. The Typicality Score compares the output distribution to a posterior distribution over classes. We show that this improves performance over the baseline in all but the segmentation task. Speculating that perhaps at the pixel level the semantic information of a pixel is less meaningful than that of class level information. Finally the new augmentation technique of DeepAugment utilizes neural networks to create augmentations on images that are radically different than the traditional geometric and camera based transformations used previously. DeepAugment improves SOTA by a significant margin while being able to be used with other augmentation schemes and demonstrates th

关键词： Information technology   Energy   Computer science  

摘要： The purpose of this research was to examine the security issues within the energy sector of our critical infrastructure against cyber threats and attack vectors. The study focused on Industrial Control Systems’ (ICS) critical role in the production, distribution, and storage of energy resource and also its failure to address outdated platforms and communication protocols that disregard foundational security principles. The study also evaluated an industry where private energy companies have displayed little interest towards threat mitigation by choosing not to adapt verified security countermeasures designed with our safety and well-being in mind. The research shows that the large adoption of Internet devices within an already vulnerable industrial area further enhances an attacker’s ability to gain access. The paper concludes with a severe concern of ongoing ICS targeted cyber-attacks based on the continued use of successful tactics that combines open source information (OSI) and malicious intent; the dateless evolution of cyber-attack vectors and our government’s inability to mandate security requirements designed with our safety and well-being in mind. Keywords: ICS, OSI, energy sector, cyber threats, attack vectors, security principles

关键词： Computer science   Artificial intelligence  

摘要： Data manipulation and visualization support data scientists' efforts to explore and understand data throughout the exploratory analysis process. Nowadays, experienced data scientists can use programming languages like SQL and R to achieve efficient and flexible analysis, and inexperienced users can easily learn and use interactive tools to accomplish simple analysis tasks. However, the lack of tools in between interactive tools and programming systems leads to a programmability gap that prevents inexperienced users from conducting expressive analysis that only users with programming experience can achieve. To help end users traverse this gap, we apply program synthesis to build tools that can synthesize programs from examples. We first introduce Falx, a visualization by example tool that lets the user create expressive visualizations using demonstrations of how a few data points are mapped to the canvas. Falx's compositional algorithm design let it synthesize both data transformation and visualization programs directly from end-to-end demonstration. We next introduce Scythe, a SQL query synthesizer that lets the user author advanced SQL queries using input-output examples. Using a language of abstract queries, Scythe can prune families of infeasible queries to achieve synthesis efficiency. To let inexperienced users distinguish synthesized complex queries, we developed a symbolic engine to compute a distinguishing input that the two queries would return different outputs. Finally, we summarize our synthesizer building experience into a framework, Kopis, that illustrates how to build an efficient relational query synthesizer using {value-preserving abstractions}. Together, these three contributions demonstrate the value of using program synthesizers to empower future data science, and offer guidance on how to build such synthesis-powered tools efficiently for new domains.

关键词： Motion pictures   Location awareness   Feature extraction   Text recognition   Convolutional neural networks   Computer science   Deep learning   Movie title extraction   multi-script   text localization   transfer learning  

摘要： Graphical texts in natural images play an important role in portraying information in multitudinous fields such as communication, education, and entertainment to name a few. Recognizing text in scene images is challenging due to the inherent complexity of the images. Text recognition in natural images involves script identification, which requires text localization. This is not trivial for natural scene images due to the presence of disparate foreground/background components. For scene images like movie posters, the challenge is more dominant. The challenges aggravate due to the presence of composite characteristics of posters like complex graphics background and the presence of different texts like a movie title, names of actors, producers, directors, and tagline. These texts have miscellaneous fonts, variations in colors, size, orientation, and textures. In this work, an M-EAST (modified EAST) model is proposed, which is based on the EAST (efficient and accurate scene text detector) model for text localization. A novel movie title extraction is thereafter used for separating the title from the extracted text pool. Finally, the title script was identified using a shallow convolutional neural network (SCNN)-based architecture to ensure functionality in low-resource environments. Experiments were performed on a dataset of movie-poster images of Tollywood, Bollywood, and Hollywood industries, and a highest accuracy of 99.82% was obtained. The system performed better than the reported techniques.

关键词： Contracts   Wireless communication   Quality of service   Base stations   Uncertainty   Numerical models   Computer science   Contract theory   incentive mechanism design   wireless edge caching   two-layer networks  

摘要： Wireless caching technologies have been proposed to relieve the transmission pressures, especially, the transmission redundancy on back-haul channels. In this paper, we consider a two-layer caching network, consisting of traditional macro-cell base station (MBS) aided back-haul channels and small-cell base stations (SBSs) aided local links. The network service provider (NSP), who is in charge of the two layers, leases its resources of the secondary layer, i.e., coverage of the SBSs, to content providers (CPs) for making extra profits and releasing pressures on the back-haul channels. At the same time, CPs will evaluate whether they are provided with proper incentives to pre-cache their files in the SBSs. Considering different quality of services (QoS) provided by the two layers as well as the economical impact of the traditional layer on the secondary layer, the NSP designs the optimal incentive mechanisms within the framework of contract theory for maximizing its own profits. First, we formulate the utility of the NSP and CPs. Then, the minimum transmission requirement, reserve price and limited resources are considered as constraints in designing the optimal contract. Also, some important properties of these constraints are analyzed to facilitate the optimal contract determination process. At last, an optimal contract determination scheme is proposed, based on which the optimal coverage set is determined first, and then the corresponding optimal prices are derived with the aid of equal cost line. Numerical results are provided to demonstrate the effectiveness of the proposed optimal contract in increasing the NSP's profits and incentivizing CPs to transmit on the secondary layer.

关键词： Computer science   Internet of Things   Spectrum allocation   Spectrum analysis   Artificial intelligence   Information sources   Decision making  

摘要： Spectrum is a public resource; yet understanding how spectrum is allocated and used is a daunting task. Usable spectrum is already fully allocated, but the demand for spectrum continues to grow and there are opportunities for utilizing spectrum in more efficient ways. Understanding how spectrum is allocated and its utilization in time and space is necessary to take advantage of these emerging opportunities. A combination of fragmented information from varied information sources, a complex regulatory environment, variability of regulations and physics by band, real-time spectrum usage dynamics, and a status quo with knowledge concentration among a few, makes understanding spectrum a considerable challenge for all stakeholders including researchers, students, policymakers, and new telecom operators. After considerable study of spectrum, its allocation, regulation, and usage, we have developed a system architecture that is a significant step towards easing the burden of understanding spectrum information. Our system architecture connects information from disparate sources and leads to a richer understanding of spectrum usage, how it is governed, and its potential for future use. Classes of information are modeled as knowledge graphs, and the interplay of knowledge graphs produces a richer set of insight and can lead to more informed decision-making. Further, we show mechanisms for connecting spectrum information with real-time observations to get a comprehensive view of spectrum usage dynamics. While focused on the United States, this work should be applicable to other spectrum contexts worldwide. This work, of considerable technical value, also has democratic value in making complex information accessible and allowing the public to determine whether spectrum, a natural resource, is being used for the public good.

关键词： Computer science   Artificial intelligence  

摘要： Fair machine learning has gained considerable interests in recent years, and has been widely adapted in every aspect of life. A multitude of formal, mathematical definitions of fairness in machine learning has been proposed in the last few years. Recently many fairness-enhancing machine learning methods have been designed to deal with various applications and scenarios. However, the research of fairness in learning over crowdsourced data and graph-structured data is largely limited. The goal of my thesis is to study algorithmic fairness in the contexts of crowd-powered data analytics and graph analytics. We have made the following research achievements. First, we focus on fair truth discovery for crowdsourcing systems. One important research problem for crowdsourcing systems is truth discovery, which aims to aggregate noisy answers contributed by the workers to obtain the correct answer (truth) of each task. However, since the collected answers are highly prone to the workers' bias, aggregating these biased answers without proper treatment will unavoidably lead to discriminatory truth discovery results. In this thesis, we address the fairness issue for truth discovery from biased crowdsourced answers. We design new algorithms that incorporate truth discovery with fairness. The algorithms estimate both worker bias and truth iteratively, and dynamically select bias to be removed from the answers during truth inference. We perform an extensive set of experiments on both synthetic and real-world crowdsourcing datasets. Experimental results demonstrate the effectiveness of our mitigation algorithms. Second, we focus on the fair classification problem under the circumstance of using applicants' social information in peer-to-peer lending. Peer-to-peer (P2P) lending marketplaces on the Web have been growing over the last decade. Since the applicants on P2P lending platforms may lack sufficient financial history for assessment, quite a few P2P lending service providers have