教学课程资源库 更多>>

关键词： Transportation   Computer science  

摘要： Improving the communication of work zone data has been the main focal point of the Work Zone Data Exchange (WZDx), which provides a standard protocol to push agency data to third party users. However, the accuracy of the work zone locations and times are difficult for many agencies to collect and maintain. Many agencies are beginning to focus on improving this data through the use of connected temporary traffic control devices such as smart arrow boards. Maintenance operations are another area that must be improved because of the short duration and the dynamic nature of the work. Currently, most maintenance operations must notify a traffic management center (TMC)of their work, but this may extend multiple miles of roadway they are working on that day, and if the work is cut short or ends early, the TMC is not always notified. To improve this data, many agencies currently equip their vehicles with AVL systems to track their snowplows. The snowplows, in most cases, also are the same vehicles used for maintenance operations such as painting, pothole repairs, etc. To minimize the work by the maintenance staff and improve the quality of work zone data, the AVL data points can be used to classify a vehicle in maintenance mode then cluster all of the surrounding vehicles to get the extents of the maintenance operation. As the vehicles move, the data will be updated in real-time to accurately communicate the extents of the maintenance operation through the agencies WZDx or ATIS system. This thesis develops an automated process that identifies maintenance activities and clusters AVL data to communicate actual maintenance operations in real-time through the WZDx.

关键词： Teacher education   Educational technology   Science education  

摘要： This study compared the learning effectiveness of two inquiry-based strategies for implementing computer simulation-based instruction for teaching science concepts: POE (Predict, Observe, and Explain) and 5E (Engage, Explore, Explain, Elaborate, and Evaluate). Participants in this study, who were pre-service teachers in science education classes, successfully completed worksheets and tests that indicated they both learned key concepts related to density as well as how to use measures of weight and volume to calculate density. Further, participants expressed appreciation of the density simulation’s ease of use and consistency of measurements. In terms of findings and implications, pre and post knowledge tests and worksheets showed no significant differences between using POE and 5E to guide use of a computer-based density simulation. This suggests that either one of these proven and accepted inquiry methods effectively enhance simulation-based learning. The choice of either one can depend on the preferences of the instructor. The POE group completed worksheets faster (38 min.) than the 5E group (41 min). No differences for the knowledge gained, problem-solving skill, and attitudes towards the simulation were shown between POE and 5E. The underlying assumption is that if pre-service teachers have a successful learning experience using science simulations in conjunction with systematic inquiry methods, then they will see the potential and value of using specific inquiry-based teaching strategies when used with computer simulation to teach science topics. This study affirmed the benefits of computer simulations and adds to the body of knowledge on the use of computer simulation plus inquiry-based teaching strategies, such as POE and 5E, to teach science. Students could learn about density concepts with either inquiry-based strategy along with computer simulation combinations, and these inquiry+simulation teaching strategy combinations should also allow science teachers

关键词： Computer science  

摘要： Interactive machine learning techniques inject domain expertise to improve or adapt models. Prior research has focused on adapting underlying algorithms and optimizing system performance, which comes at the expense of user experience. This dissertation advances our understanding of how to design for human-machine collaboration—improving both user experience and system performance—through four studies of end users' experience, perceptions, and behaviors with interactive machine learning systems. In particular, we focus on two critical aspects of interactive machine learning: how systems explain themselves to users (transparency) and how users provide feedback or guide systems (control). We first explored how explanations shape users' experience of a simple text classifier with or without the ability to provide feedback to it. Users were frustrated when given explanations without means for feedback and expected model improvement over time even in the absence of feedback. To explore transparency and control in the context of more complex models and subjective tasks, we chose an unsupervised machine learning case, topic modeling. First, we developed a novel topic visualization technique and compared it against common topic representations (e.g., word lists) for interpretability. While users quickly understood topics with simple word lists, our visualization exposed phrases that other representations obscured. Next, we developed a novel, "human-centered'' interactive topic modeling system supporting users' desired control mechanisms. A formative user study with this system identified two aspects of control exposed by transparency: adherence, or whether models incorporate user feedback as expected, and stability, or whether other unexpected model updates occur. Finally, we further studied adherence and stability by comparing user experience across three interactive topic modeling approaches. These approaches incorporate input differently, resulting in varied adherence, st

关键词： Computational methods   computer science   engineering science and mechanics   machine learning   sustainable energy   wind power  

摘要： This work presents a novel investigation on the nowcasting prediction of wind speed for three sites in Bahia, Brazil. For this, it was applied the computational intelligence by supervised machine teaming using different artificial neural network technique, which was trained, validated, and tested using time series are derived from measurements that are acquired in towers equipped with anemometers at heights of 100.0, 120.0 and 150.0 m. To define the most efficient ANN, different topologies were tested using MLP and RNN, applying Wavelet packet decomposition (bior, coif, db, dmey, rbior, sym). The best statistical analysis was RNN + discrete Meyer wavelet.

关键词： Geology   Computer engineering   Computer science  

摘要： Previous methods to determine the origin of natural gases consist mostly of analyzing combinations of three empirical binary plots with molecular ratio C1/(C2+C3) and isotopic values δ13C-C1, δ12H-C1, and δ13C-CO2. Using these diagrams, geochemists distinguish five origins of natural gas: abiotic, thermogenic, primary microbial from CO2 reduction primary microbial from methyl-type fermentation and secondary microbial. However, the genetic fields on these diagrams partially overlap, making the interpretation of some samples ambiguous. Here, I integrate supervised machine learning to improve these methods and create a tool that dramatically increases the accuracy, certainty and speed of classifying a natural gas sample. With data science and software engineering, I utilized a dataset of 27,853 natural gas samples to create and deploy a random forest natural gas classifier tool that uses the same geochemical parameters as the binary diagrams. This web tool that can automatically classify the origin of natural gas at an accuracy of over 97% and provide calculated certainty of each classification.

关键词： Computer science  

摘要： Generative Adversarial Networks (GANs) have been under the spotlight in the machine learning field for a few years. Especially, the power that learns a data distribution in an unsupervised fashion leads GANs to be applied to various applications such as page generation, image style transformation, image attribution manipulation, and similar domains in computer vision. Despite the huge success of GANs, the difficult and unstable training process still limits the applications of GANs in the real world. Mode collapse is a well-known byproduct of unstable GAN training. We propose to improve the sample diversity of a pre-trained class- conditional generator by modifying its class embeddings in the direction of maximizing the log probability outputs of a classifier pre-trained on the same dataset. We improved the sample diversity of state-of-the-art ImageNet BigGANs at both 128 × 128 and 256 × 256 resolutions. By replacing the embeddings, We can also synthesize plausible images for Places365 using a BigGAN pre-trained on Image: Net.

关键词： Statistics   Computer science  

摘要： Energy-based models are a powerful and flexible tool for studying emergent properties in systems with many interacting components. Energy functions of complex systems are highly non-convex because the local modes encode the rich variety of probable phenomena. This work focuses on energy models of image signals, which are treated as complex systems of interacting pixels that contribute to the emergent appearance of the whole image. By adapting classical Maximum Likelihood learning to the parametric family of ConvNet functions, one can learn energy-based models capable of realistic image synthesis. The observed behaviors of ML learning with ConvNet potentials are surprising and not well understood despite widespread use of the potential in recent studies. This work rigorously diagnoses the learning process to correct systematic problems with steady-state distributions of learned potentials while revealing an unexpected non-convergent outcome of learning. The true value of the energy-based model lies in the structure of the energy landscape. The energy-based model encodes explicit quantitative relations between local modes in landscape energy barriers. Concepts in the training data form macroscopic energy basins with many shallow local modes. Powered by a new Markov Chain Monte Carlo algorithm that efficiently detects macroscopic energy structures, energy landscape mapping experiments are conducted to discover meaningful image concepts. The mapping framework is a powerful tool for unsupervised clustering based on geodesic distances in a learned potential.

关键词： Computer engineering   Computer science   manycore processor   parallel algorithms   PRAM   XMT  

摘要： This dissertation demonstrates the possibility of obtaining strong speedups for a variety of parallel applications versus the best serial and parallel implementations on commodity platforms. These results were obtained using the PRAM-inspired Explicit Multi-Threading (XMT) many-core computing platform, which is designed to efficiently support execution of both serial and parallel code and switching between the two. Biconnectivity: For finding the biconnected components of a graph, we demonstrate speedups of 9x to 33x on XMT relative to the best serial algorithm using a relatively modest silicon budget. Further evidence suggests that speedups of 21x to 48x are possible. For graph connectivity, we demonstrate that XMT outperforms two contemporary NVIDIA GPUs of similar or greater silicon area. Prior studies of parallel biconnectivity algorithms achieved at most a 4x speedup, but we could not find biconnectivity code for GPUs to compare biconnectivity against them. Triconnectivity: We present a parallel solution to the problem of determining the triconnected components of an undirected graph. We obtain significant speedups on XMT over the only published optimal (linear-time) serial implementation of a triconnected components algorithm running on a modern CPU. To our knowledge, no other parallel implementation of a triconnected components algorithm has been published for any platform. Burrows-Wheeler compression: We present novel work-optimal parallel algorithms for Burrows-Wheeler compression and decompression of strings over a constant alphabet and their empirical evaluation. To validate these theoretical algorithms, we implement them on XMT and show speedups of up to 25x for compression, and 13x for decompression, versus bzip2, the de facto standard implementation of Burrows-Wheeler compression. Fast Fourier transform (FFT): Using FFT as an example, we examine the impact that adoption of some enabling technologies, including silicon photonics, would have on the perfo

关键词： Computer science  

摘要： Back in 1984 when information sharing barely existed and computing resources were limited, a space trading simulator called Elite created a whole new genre of procedural generation of game content. The object of the game was to fly from solar system to solar system, visiting planets, buying and selling goods. All galaxies (eight of them at the time) with 256 stars each were procedurally generated with economy, population, government, etc. Procedurally generated game content could result in large amounts of game content requiring fewer computing resources. In today’s world where we have access to a large amount of computing resources, procedurally generated game worlds could mean unique, unlimited experiences for the players. Modern graphics processing power also allows for the visualization of procedurally generated game content in way that was not possible in the 80s. The purpose of my research was to investigate some of these techniques and algorithms and combine them in order to procedurally generate interesting and detailed game worlds and content in order to enhance the player’s experience. The research was centered around exploring mature and effective game developer toolsets and techniques like pseudo random number generators, coherent noise algorithms and modern 3D graphic engines. Integration of these techniques was possible through solid and creative software design and architecture patterns. As I guide you through the approach taken to the learning and application of these techniques, it will be clear that it is possible for these algorithms to exist in harmony as a form of a diverse software ecosystem. Moreover, it will become clear that the goals of this research were achieved and the blueprints and content of an enhanced game world where generated procedurally by computing algorithms.

关键词： Engineering   Robotics   Computer science  

摘要： Simultaneous localization and mapping (SLAM) is a class of techniques that allow robots to navigate unknown environments using onboard sensors. With inexpensive commercial cameras as the primary sensor, visual SLAM has become an important and widely used approach to enabling mobile robot autonomy. However, traditional visual SLAM algorithms use only a fraction of the information available from conventional cameras: in addition to the basic geometric cues typically used in visual SLAM, colour images encode information about the camera itself, environmental illumination, surface materials, vehicle motion, and other factors influencing the image formation process. Moreover, visual localization performance degrades quickly in long-term deployments due to environmental appearance changes caused by lighting, weather, or seasonal effects. This is especially problematic when continuous metric localization is required to drive vision-in-the-loop systems such as autonomous route following. This thesis explores several novel approaches to exploiting additional information from cameras to improve the accuracy and reliability of metric visual SLAM algorithms in short- and long-term deployments. First, we develop a technique for reducing drift error in visual odometry (VO) by estimating the position of a known light source such as the sun using indirect illumination cues available from existing image streams. We build and evaluate hand-engineered and learned models for single-image sun detection and achieve significant reductions in drift error over 30~km of driving in urban and planetary analogue environments. Second, we explore deep image-to-image translation as a means of improving metric visual localization under time-varying illumination. Using images captured under different illumination conditions in a common environment, we demonstrate that localization accuracy and reliability can be substantially improved by learning a many-to-one mapping to a user-selected canonical ap