教学课程资源库 更多>>

关键词： Computer science  

摘要： Content creation can be broadly defined as a way of conveying thoughts and expressing ideas through some medium such as speech, text or any of various arts. The general goal of content creation is to generate contents that make the information accessible and understandable to audiences. In recent years, with the rapid progress in artificial intelligence, it has become an inevitable fact that the future of content creation is a powerful blend of machine technology and human creativity. However, it remains extremely challenging for machines to truly emulate what a content creator does. Therefore, instead of attempting to take over the role of professional content creators, we aim at (1) how to shorten the gap between professional content creators and general users with the help of machines, and (2) how to leverage machines to facilitate the creation process. In this work, we propose efficient algorithms based on generative models to tackle several content creation tasks that are originally time- and money-consuming. First, we address the problem of image-to-image translation. We propose a disentangled representation image-to-image translation (DRIT) framework to perform diverse translation without paired training data. Our model disentangles images into a domain-invariant content space and a domain-specific attribute space to enable diverse translation. Furthermore, to improve the diversity of the generated images, we propose a simple yet effective model-seeking regularization term. The proposed regularization term can serve as a plug-in term to various conditional generation tasks. Second, we address the music-to-dance translation task. Given an input music clip, we aim to generate a corresponding dancing sequence. We propose a synthesis-by-analysis learning framework. We first learn to perform basic movement, then learn how to combine basic. movements based on input music. The generated dancing sequences are consistent to the input music clips in terms of music styl

关键词： Computer engineering   Computer science  

摘要： With mobile systems evermore ubiquitous, individual users often own multiple mobile systems and groups of users often have many mobile systems at their disposal. As a result, there is a growing demand for multi-mobile computing, the ability to combine the functionality of multiple mobile systems into a more capable one. However, there are several key challenges. First, mobile systems are highly heterogeneous with different software and hardware, each with their own interfaces and data formats. Second, there are no effective ways to allow users to easily and dynamically compose together multiple mobile systems for the quick interactions that typically take place with mobile systems. Finally, there is a lack of system infrastructure to allow existing apps to make use of multiple mobile systems, or to enable developers to write new multi-mobile aware apps. My thesis is that higher-level abstractions of mobile operating systems can be reused to combine heterogeneous mobile systems into a more capable one and enable existing and new apps to provide new functionality across multiple mobile systems. First, we present M2, a system for multi-mobile computing that enables existing unmodified mobile apps to share and combine multiple devices, including cameras, displays, speakers, microphones, sensors, GPS, and input. To support heterogeneous devices, M2 introduces a new data-centric approach that leverages higher-level device abstractions and hardware acceleration to efficiently share device data, not API calls. M2 introduces device transformation, a new technique to mix and match heterogeneous devices, enabling, for example, existing apps to leverage a single larger display fused from multiple displays for better viewing, or use a Nintendo Wii-like gaming experience by translating accelerometer to touchscreen input. We have implemented M2 and show that it operates across heterogeneous systems, including multiple versions of Android and iOS, and can run existing apps across m

关键词： Computer science  

摘要： In the era of Web 2.0, people express their opinion, feelings and thoughts about topics including political and cultural events, natural disasters, products and services, through mediums such as blogs, forums, and micro-blogs, like Twitter. Also, large amount of text is generated through e-mail which contains the writer's feeling or opinion; for instance, customer care service e-mail. The texts generated through such platforms are a rich source of data, which can be mined in order to gain useful information about user opinion or feelings. Sentiment Analysis identifies and extracts information about the attitude of a speaker or writer on a subject, topic, polarity, or emotion in a document. Sentiments can be extracted from sources such as speech, music, and facial expression. Due to rich source of information available in the form of text data, we focus on sentiment analysis and emotion mining from text. We further discover Actionable Patterns from these sentiments, which suggest ways to alter the user's emotion to a more positive or desirable state. Little work has been done on extracting Actionable Recommendations from sentiments. We contribute to the solution of this challenging problem by applying machine learning methods such as decision forest, and support vector machines for emotion classification, and Action Rules Mining for Emotion altering recommendations. We experiment with live streaming Twitter Data, Student Evaluations data, Business Net Promotor Score data. Results show high accuracy for Emotion Detection, and successful discovery of Actionable Recommendations for more positive user sentiment. Applications of this work include: marketing, sale predictions, political surveys, health care, student-faculty culture, e-learning platforms, and social networks.

关键词： Task analysis   Bibliographies   Licenses   Periodic structures   Redundancy   Computer science   Coherence   Cross-document structure theory   discourse relations   multi-article summarization   rhetorical structure theory  

摘要： Over the last decade, discourse relations, also referred to as rhetorical or coherence relations, have been used to improve a range of natural language processing applications. Researchers have devised several theories, including rhetorical structure theory and cross-document structure theory, to examine relations between generic text units in single and multiple documents, respectively. In this paper, we propose a cross-article structure theory (CAST), that extends the benefit of discourse relations to multi-scientific article applications. It is based on the rhetorical structure theory (RST) and the cross-document structure theory (CST). The insight that underpins CAST is to consider both intra-section and cross-section relations. At the outset, these relations are classified based on the structural features of the article (that is, their appearance within each section type) and then the relations between text portions across multiple articles are classified. The practicality of the theory is showcased by solving a problem that consists to identify the types of relations which exist between each pair of sentences in related sections of different articles. A CAST bank was created and the k-nearest neighbors algorithm was used to develop two classifiers based on CAST and CST, respectively. The performance results obtained markedly demonstrate the role of the specific relations to scientific articles in CAST. Other applications of CAST could address the redundancy and readability problems, which represent main issues for different tasks, such as the summarization of multiple articles.

关键词： Applied mathematics   Artificial intelligence   Computer science  

摘要： We study the diffusivity of random walks with transition probabilities depending on the number of consecutive traversals of the last traversed edge, the so called senile reinforced random walk (SeRW). In one dimension, the walk is known to be sub-diffusive with identity reinforcement function. We perturb the model by introducing a small probability δ of escaping the last traversed edge at each step. The perturbed SeRW model is diffusive for any δ > 0, with enhanced diffusivity much greater than O(δ2)) in the small δ regime. We further study stochastically perturbed SeRW models by having the last edge escape probability of the form δξn with ξn's being independent random variables. Enhanced diffusivity in such models are logarithmically close to the so called residual diffusivity (positive in the zero δ limit). Finally, we generalize our results to higher dimensions where the unperturbed model is already diffusive. Regularization of deep neural networks (DNN's) is one of the effective complexity reduction methods to improve efficiency and generalizability. We consider the problem of regularizing a one hidden layer convolutional neural network with ReLU activation function via gradient descent under sparsity promoting penalties. It is known that when the input data is Gaussian distributed, no-overlap networks (without penalties) in regression problems with ground truth can be learned in polynomial time at high probability. We propose a Relaxed Variable Splitting Method (RVSM), integrating thresholding and gradient descent to overcome the non-smoothness in the associated loss function. The sparsity in network weight is realized during the optimization (training) process. We prove that under ℓ1; ℓ0; and transformed-ℓ1 penalties, no-overlap networks can be learned with high probability, and the iterative weights converge to a global limit which is a transformation of the true weight under a novel thresholding operation. Numerical experiments confirm theoretical finding, a

关键词： Information science   Artificial intelligence   Computer science  

摘要： Spatial data have tremendous value and are necessary components in many important societal applications. In recent years, our world has been witnessing a revolution brought by spatial technologies (e.g., Google Maps, Waze, Uber, Lyft, Grubhub, Lime, autonomous driving). According to a McKinsey Global Institute report, location data will generate about $600 billion annual revenue by 2020 with applications in energy, health, retail, etc. The world's economy also heavily relies on location and time data from over 2 billion GPS receivers, and these data are essential to many applications such as banks, airlines, police, emergency services, and telecommunications. Meanwhile, new types of spatial data are emerging at unprecedented scales and varieties (e.g., 25GB/hour per connected vehicle, 47.7PB per year by NASA by 2022). While spatial data are critical, valuable and collected at massive scales, they pose great challenges to traditional artificial intelligence (AI) techniques when applied to important societal problems. This thesis addresses three of these challenges. First, spatial data (e.g., crime or disease distribution, air quality) are often directly linked to our lived environments. As a result, decisions made on such data tend to have direct impacts on the life of citizens, and thus require statistical robustness to avoid errors which can have high economic and social costs (e.g., false alarm of a crime hotspot). Second, spatial data exhibit interdependency and variability, violating the common i.i.d. (identically and independently distributed) assumption in traditional statistics. This introduces new challenges to traditional optimization problems where spatial interdependency between nearby locations is often neglected and understudied (e.g., spatial contiguity required in land allocation). Finally, data and domain knowledge gaps are common in geospatial problems. For example, while Earth observation imagery is available in the tens of petabytes, there is very

关键词： Computer engineering   Electrical engineering   Computer science  

摘要： Data centers are heavily dependent upon transceivers for the exchange of data. This work analyzes various components of a transceiver ((de-)multiplexer, modulator, and laser) and evaluates how to exchange or enhance performance by incorporating Bragg based structures. First, it is shown an apodized 4-port contra-directional coupler can perform wavelength division multiplexing (WDM) at O-band wavelengths in a semi-dense channel grid. The devices are concatenated to create a 16-channel system where normally a 4-channel system is utilized. The concept of WDM is extended to mode division multiplexing, where the orthogonal spatial modes of a waveguide are excited and controlled. It is demonstrated Bragg devices can excite the TE01 mode while also controlling it in a resonant format. Next, the apodized 4-port structure is used to demonstrate an arbitrary and tunable spectral shaping device. By showcasing a wide array of intuitive spectral building blocks it is shown that multiple functions, such as WDM or modulation, can be achieved with a single device. Finally, a laser is created by coupling a reflective semiconductor optical amplifier (RSOA) to a Bragg based structure. The Bragg structure completes the laser cavity by operating as a mirror and can be tuned in wavelength and in intensity, where the latter has a GHz response time. Combining these devices suggests a transceiver can largely be composed of Bragg based structures, which has the potential to simplify the design of future transceivers.

关键词： Computer science  

关键词： Transportation   System science   Computer science  

摘要： Fuel efficiency is an ever present problem in today's modern world. The United States in particular is in need of a solution to lowering greenhouse gas emissions caused by transit and freight across its spread out cities. In fact in the United States the average commute time of an individual is 26 minutes, meaning that round trip people are driving about an hour every day, to and from work. But that gas consumption is pittiling compared to that of the freight industry. Heavy Duty Vehicles (HDVs) commonly known as semi trucks, account for three quarters of US freight emissions and 7.5% of total US greenhouse gas emissions. But this can be cut down considerably with the implementation of platooning. Platooning is when multiple vehicles follow in close distance to reduce aerodynamic drag, causing significant fuel savings. In this paper, we will go over an algorithm to help vehicles join with already formed platoons on the road, increasing their fuel efficiency and therefore saving cost to the driver in addition to lessening the negative effect on the planet. This will be done using a modified A* algorithm. The base weight of zero will be the amount of gas the vehicle would consume taking the shortest path that google maps recommends, alone, with no platooning. Paths may end up with negative weights due to the fuel savings caused by joining existing platoons during the vehicles’ travel. The algorithm will have access to a map of the roadways and the GPS data of nearby platoons. It will then perform a cost-benefit analysis to determine if the fuel savings from joining a platoon will outweigh the cost of going outside of its original path in order to join the other vehicles.

关键词： Computer science  

摘要： In healthcare, a large number of medical data has emerged. To effectively use these data to improve healthcare outcomes, clinicians need to identify the relevant measures and apply the correct analysis methods for the type of data at hand. In this dissertation, we present various machine learning (ML) and data mining (DM) methods that could be applied to the type of data sets that are available in the healthcare area. The first part of the dissertation investigates DM methods on healthcare or medical data to find significant information in the form of rules. Class association rule mining, a variant of association rule mining, was used to obtain the rules with some targeted items or class labels. These rules can be used to improve public awareness of different cancer symptoms and could also be useful to initiate prevention strategies. In the second part of the thesis, ML techniques have been applied in healthcare or medical data to build a predictive model. Three different classification techniques on a real-world breast cancer risk factor data set have been investigated. Due to the imbalance characteristics of the data set various resampling methods were used before applying the classifiers. It is shown that there was a significant improvement in performance when applying a resampling technique as compared to applying no resampling technique. Moreover, super learning technique that uses multiple base learners, have been investigated to boost the performance of classification models. Two different forms of super learner have been investigated - the first one uses two base learners while the second one uses three base learners. The models were then evaluated against well-known benchmark data sets related to the healthcare domain and the results showed that the SL model performs better than the individual classifier and the baseline ensemble. Finally, we assessed cancer-relevant genes of prostate cancer with the most significant correlations with the clinical outcome o