教学课程资源库 更多>>

关键词： Computer engineering   Electrical engineering   Computer science  

摘要： Power and energy have become primary design constraints for modern processors. Power no longer scales with transistor channel length, causing power density to increase. Continued transistor scaling will require prohibitively expensive cooling technologies. This is particularly an issue in data centers, where the cost of power and cooling has a major impact on the total cost of ownership. Data centers have grown and expanded rapidly due to the advent of cloud computing and interactive web services. The unabated migration of compute to data centers means future data centers will continue to require more compute which can be cooled in a cost-effective way. As a result, architects have looked to manycore processors due to their performance on parallel workloads and energy efficiency benefits. As opposed to conventional multicore data center processors which utilize a few large out-of-order cores optimized for instruction level parallelism, manycores utilize many smaller cores optimized for throughput. The sea of smaller cores work in parallel on a problem, each working on a smaller portion of the overall problem. As a result, each core can run slower, and the same amount of work can be completed in the same amount of time using less energy. The focus of this thesis is to understand the power and energy characteristics of manycore processors and improve their already superior energy efficiency. Execution Drafting (ExecD) is a microarchitectural mechanism that exploits commonality between identical or similar programs or threads in data centers to improve the energy efficiency of fine-grained multithreaded cores. When similar or identical threads are scheduled to the same core, a thread synchronization mechanism that attempts to synchronize the threads in time to the same instruction takes over the thread scheduling decision. Once synchronized, identical instructions from different threads are consecutively issued down the pipeline, leading to reduced activity factors on

关键词： Computer science   bibliometrics   institutions   web of science   citations   ranking  

摘要： Computer Science degrees are very popular currently among institutions worldwide. The proliferation of these programs in different universities has led to the creation of rankings for classifying programs according to their prestige and quality. However, these rankings do not specify the quality of research. This study develops a bibliometric overview of all the journals that are currently indexed in the Web of Science (WoS) database in any of the seven categories connected to Computer Science research. These categories include Artificial Intelligence, Cybernetics, Hardware and Architecture, Information Systems, Interdisciplinary Applications, Software Engineering and Theory and Methods. This study aims to identify the leading institutions over the last 25 years (1991-2015) in each area selected according to a wide range of bibliometric indicators. The results indicate that American universities are the most influential in Computer Science research. This study concludes that Computer Science traverses many institutions.

关键词： Applied mathematics   Computer science  

摘要： We present an extension to the Material Point Method (MPM) for simulating elastic objects with various co-dimensions like hair (1D), thin shells (2D), and volumetric objects (3D). We simulate thin shells with frictional contact using a combination of MPM and subdivision finite elements. The shell kinematics are assumed to follow a continuum shell model which is decomposed into a Kirchhoff-Love motion that rotates the mid-surface normals followed by shearing and compression/extension of the material along the mid-surface normal. We use this decomposition to design an elastoplastic constitutive model to resolve frictional contact by decoupling resistance to contact and shearing from the bending resistance components of stress. We show that by resolving frictional contact with a continuum approach, our hybrid Lagrangian/Eulerian approach is capable of simulating challenging shell contact scenarios with hundreds of thousands to millions of degrees of freedom. Furthermore our technique naturally couples with other traditional MPM methods for simulating granular materials. Without the need for collision detection or resolution, our method runs in a few minutes per frame in these high resolution examples. For the simulation of hair and volumetric elastic objects, we utilize a Lagrangian mesh for internal force computation and an Eulerian mesh for self collision as well as coupling with external materials. While the updated Lagrangian discretization where the Eulerian grid degrees of freedom are used to take variations of the potential energy is effective in simulating thin shells, its frictional contact response strategy does not generalize to volumetric objects. Therefore, we develop a hybrid approach that retains Lagrangian degrees of freedom while still allowing for natural coupling with other materials simulated with traditional MPM. We demonstrate the efficacy of our technique with examples that involve elastic soft tissues coupled with kinematic skeletons, extreme de

关键词： Computer science  

摘要： Research on time-series classification has garnered importance among practitioners in the data mining community. A major reason behind the ever-increasing interest among data-miners is the plethora of time-series data available from a wide range of real-life domains. Temporal-ordered data from a variety of sensor-based domains such as wearable devices, smart homes, industrial monitoring, medical diagnosis, etc. provide classification challenges more akin to real-world scenarios. Thus, building more robust time-series classification models is imperative. One group of popular models focuses on identifying short discriminative temporal patterns (subsequences) from the time-series for classification. These temporal subsequences, known as shapelets, are local patterns that can be used to uniquely identify the target class of a time-series instance. In this dissertation, I explore two real-world challenges pertaining to shapelet based time-series classification models and provide solutions to mitigate those challenges. In the first challenge, the problem of cost-sensitive learning in time-series classification is explored. First, the problem of highly imbalanced time-series classification using shapelets is investigated. The current state-of-the-art approach learns generalized shapelets along with weights of the classification hyperplane via a classical cost-insensitive loss function. Cost-insensitive loss functions tend to treat different misclassification errors equally, and thus, models are usually biased towards examples of the majority class. In this research, the generalized shapelets learning framework is extended and a cost-sensitive learning model is proposed. Instead of incorporating the misclassification cost as prior knowledge, as was done by other published methods, a constrained optimization problem was formulated to learn the unknown misclassification costs along with the shapelets and their weights. Secondly, I focus on the problem of cost-sensitive early

关键词： Computer science  

摘要： In this thesis, we study basic lower bound questions in communication complexity, data structures and depth-2 threshold circuits, and prove lower bounds in these models by devising new techniques in information theory, algebra and combinatorics. Communication Complexity: A central open problem in communication complexity is to determine whether the messages exchanged by two parties can be compressed if we know that the amount of information revealed by the parties about their inputs is small. We consider the compression question when the information revealed by one of the parties is much less than the information revealed by the other. In this setting, we prove two new improved compression schemes. Data Structures: Our contribution to data structure lower bounds is threefold: (a) Consider the Vector-Matrix-Vector problem, in which the data structure stores a \sqrt{n} \times \sqrt{n} bit matrix and provides an algorithm to compute uMv (mod 2) for \sqrt{n}-bit vectors u, v. We prove new static data structure lower bounds for this problem, which improve upon the previous work of Chattopadhyay, Kouck\'{y}, Loff, and Mukhopadhyay by a factor of log n. Our proof uses a new technique by combining the discrepancy method from communication complexity with a modification of cell sampling. This technique turns out to be more general, and can be used to prove strong lower bounds for data structures that err and have a binary query output. (b) We show new connections between systematic linear data structures, linear data structures and matrix rigidity. Specifically, we prove the equivalence between systematic linear data structures and set rigidity, a relaxation of matrix rigidity that was defined by Alon, Panigrahy and Yekhanin. This equivalence not only sheds light on the difficulty of proving strong lower bounds against data structures but also suggests candidate rigid sets from data structures. We also use this equivalence to relate linear data structures and rigidity. (c) W

关键词： Statistics   Artificial intelligence   Computer science  

摘要： Robustness to adversarial perturbation has become an extremely important criterion for applications of deep neural networks in many security-sensitive domains such as spam detection, fraud detection, criminal justice, malware detection, and financial markets. Thus, for a given model in security sensitive applications, the evaluation of its robustness to adversarial attacks are more and more crucial. In this thesis, we present a probabilistic framework for generating adversarial attacks on discrete data. Based on this framework, we derive a perturbation-based method, Greedy Attack, and a scalable learning-based method, Gumbel Attack, that illustrate various trade-offs in the design of attacks. We also propose a meta algorithm called BOSH-attack that boost the performance of current decision-based attack algorithms under the hard-label black-box setting. Finally we develop a detection method to detect maliciously crafted adversarial examples via tools in model interpretation.

摘要： On August 31, 1970, an experiment began that continues to this day. The first chess tournament for computers was held as part of the Association for Computing Machinery's (ACM's) National Conference. The interest generated was tremendous, leading to ACM sponsoring an annual event until 1994. Chess competitions continue to this day, allowing for 50 years of data on the growth of artificial intelligence capabilities in this domain. During this period, program ratings have soared from roughly 1400 in 1970 to over 3500 today. The 1970 event was the first continuous competition in computer science history, and it represents the longest ongoing experiment in computing history. 1 Some paragraphs of the text have been taken from Man vs. Machine: Challenging Human Supremacy at Chess by Karsten Müller and Jonathan Schaeffer [2018]. [ABSTRACT FROM AUTHOR]

关键词： Field of study trend   citation trend   clustering   computer science  

摘要： The volume and diversity of scientific literature are escalating every day and millions of new scientific articles are published every year. Researchers work and publish in their respective fields of interest. A major portion of the scientific community publishing in the same field of interest forms a trend in the field which could be deemed as relatively more popular than other trends. A pioneering researcher picks a field of interest by depending upon its popularity. This may have a positive impact on the acceptance of a study or high count of citations in future. This study identifies how significant it is to follow a research trend and the impact of the field of study (FoS) trend on research paper citations. For this purpose, we have chosen the field of Computer Science and Microsoft Academic Graph dataset from the 2007-2015 time period. In the dataset, every paper has a list of fields of study. The FoS provided in MAG is systematized hierarchically into 4 levels;level-0 - level-3. In this study, we apply the clustering technique to the FoS and citations pattern separately. Likewise, we also analyze how papers following a FoS trend, gain citations over the time. We also introduce a novel method Field of Study Multigraph (FoM) using graph centrality measures degree, betweenness and closeness to analyze the FoS trend, citation trend, and the relation between research areas in scientific articles from the domain of Computer Science. The experimental results show that the FoS has a certain impact on citation count. Furthermore, the results depicts that if papers belong to the same FoS, then there are 66% of the chances of having a similar citation pattern and that they have the same citation trend as they also achieved a high correlation value. This proves that a FoS has a certain impact on the citation count of a paper and researchers should contemplate the FoS trend before selecting a particular research area.

关键词： Artificial intelligence   Computer science  

摘要： Alongside the ever-growing influence of convolutional neural networks(CNN) in the image classification task, its capability for working robustly with noisy images will become more critical. A practical classification system needs to deal with noisy images because the inputs from the real-world environment that such a system will have to operate in can be noisy. Inspired from findings in Neuroscience, this work investigates whether Gabor filters can benefit the noise robustness of CNNs. It is widely known that responses from V1 cells in the early visual cortex can be modeled with Gabor functions. Visual pipeline in mammalian brains has to learn how to extract salient features from a noisy environment throughout its lifetime. Thus, we can suspect that these Gabor filter-like responses from V1 cells contribute to the noise robustness of visual perception. Given the assumption that there are similarities between CNNs and human visual perception, we investigate whether using Gabor filters can enhance CNN's capability for classifying images infused with noise.

关键词： 计算机科学与技术   教学问题   就业形势   适应  

摘要： 在社会发展形势的变化下,社会对计算机人才的要求越来越高,为了适应这一局势,教育行业也在进行着如火如荼的改革,计算机科学与技术教学也是如此。针对当前计算机科学与技术教学的问题进行分析,探讨计算机科学与技术教学如何适应就业形势。