检索结果-内蒙古大学图书馆

Unusual structures inherent in point pattern data predict colon cancer patient survival

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Jones-Todd, By Charlotte M. Caie, Peter Illian, Janine Stevenson, Ben C. Savage, Anne Harrison, David J. Bown, James L. School of Mathematics and Statistics University of St.Andrews Creem School of Medicine University of St.Andrews Department of Statistics University of Auckland School of Arts Abertay University Media and Computer Games School of Science Abertay University Engineering and Technology

Cancer patient diagnosis and prognosis is informed by assessment of morphological properties observed in patient tissue. Pathologists normally carry out this assessment, yet advances in computational image analysis provide opportunities for quantitative assessment of tissue. A key aspect of that quantitative assessment is the development of algorithms able to link image data to patient survival. Here, we develop a point process methodology able to describe patterns in cell distribution within cancerous tissue samples. In particular, we consider the Palm intensities of two Neyman Scott point processes, and a void process developed herein to reflect the spatial patterning of the cells. An approximate-likelihood technique is taken in order to fit point process models to patient data and the predictive performance of each model is determined. We demonstrate that based solely on the spatial arrangement of cells we are able to predict patient survival. Copyright © 2017, The Authors. All rights reserved.

关键词： Hospital data processing

Optimal power control for transmitting correlated sources with energy harvesting constraints

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Dong, Yunquan Chen, Zhi Wang, Jian Shim, Byonghyo School of Electronic and Information Engineering Nanjing University of Information Science and Technology Nanjing210044 China Department of Electronic and Computer Engineering University of Waterloo WaterlooONN2L3G1 Canada Institute of New Media and Communications Department of Electrical and Computer Engineering Seoul National University Seoul151744 Korea Republic of

We investigate the weighted-sum distortion minimization problem in transmitting two correlated Gaussian sources over Gaussian channels using two energy harvesting nodes. To this end, we develop offline and online power control policies to optimize the transmit power of the two nodes. In the offline case, we cast the problem as a convex optimization and investigate the structure of the optimal solution. We also develop a generalized water-filling based power allocation algorithm to obtain the optimal solution efficiently. For the online case, we quantify the distortion of the system using a cost function and show that the expected cost equals the expected weightedsum distortion. Based on Banach's fixed point theorem, we further propose a geometrically converging algorithm to find the minimum cost via simple iterations. Simulation results show that our online power control outperforms the greedy power control where each node uses all the available energy in each slot and performs close to that of the proposed offline power control. Moreover, the performance of our offline power control almost coincides with the performance limit of the system. Copyright © 2017, The Authors. All rights reserved.

关键词： Energy harvesting

Virtual worlds as proxy for multi-object tracking analysis

学校读者我要写书评

暂无评论

Virtual worlds as proxy for multi-object tracking analysis

2016 IEEE Conference on computer Vision and Pattern Recognition, CVPR 2016

作者： Gaidon, Adrien Wang, Qiao Cabon, Yohann Vig, Eleonora Computer Vision Group Xerox Research Center Europe France School of Electrical Computer and Energy Engineering School of Arts Media and Engineering Arizona State University United States German Aerospace Center Germany

ISBN: (纸本)9781467388511

Modern computer vision algorithms typically require expensive data acquisition and accurate manual labeling. In this work, we instead leverage the recent progress in computer graphics to generate fully labeled, dynamic, and photo-realistic proxy virtual worlds. We propose an efficient real-to-virtual world cloning method, and validate our approach by building and publicly releasing a new video dataset, called "Virtual KITTI"1, automatically labeled with accurate ground truth for object detection, tracking, scene and instance segmentation, depth, and optical flow. We provide quantitative experimental evidence suggesting that (i) modern deep learning algorithms pre-trained on real data behave similarly in real and virtual worlds, and (ii) pre-training on virtual data improves performance. As the gap between real and virtual worlds is small, virtual worlds enable measuring the impact of various weather and imaging conditions on recognition performance, all other things being equal. We show these factors may affect drastically otherwise high-performing deep models for tracking.

关键词： computer vision

Automatic diacritics restoration for modern standard Arabic text

学校读者我要写书评

暂无评论

Automatic diacritics restoration for modern standard Arabic ...

2016 IEEE Symposium on computer Applications and Industrial Electronics, ISCAIE 2016

作者： Zayyan, Ayman A. Elmahdy, Mohamed Binti Husni, Husniza Al Ja'Am, Jihad M. Department of Computer Science and Engineering Faulty of Media Engineering and Technology Qatar University Qatar German University in Cairo State of Qatar Cairo Egypt School of Computing College of Arts and Sciences Qatar

ISBN: (纸本)9781509015436

In this paper, the problem of missing diacritic marks in most of Arabic written resources is investigated. Our aim is to implement a scalable and extensible platform to automatically restore missing diacritic marks for Modern Standard Arabic text. Different rule-based and statistical techniques are proposed. These include: morphological analyzer-based, maximum likelihood estimate, and statistical n-gram models. Diacritization accuracy of each technique was evaluated based on Diacritic Error Rate (DER) and Word Error Rate (WER). The proposed platform includes helper tools for text preprocessing and encoding conversion. It yielded a WER of 7.1% and DER of 3.9%. When the case ending was ignored, the platform yielded a WER and DER of 5.1% and 2.7%, respectively. © 2016 IEEE.

关键词： Text processing

Procedural content generation via machine learning (PCGML)

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Summerville, Adam Snodgrass, Sam Guzdial, Matthew Holmgård, Christoffer Hoover, Amy K. Isaksen, Aaron Nealen, Andy Togelius, Julian Department of Computational Media University of California Santa CruzCA95064 United States College of Computing and Informatics Drexel University PhiladelpiaPA19104 United States School of Electrical and Computer Engineering Georgia Institute of Technology AtlantaGA30332 United States Duck and Cover Games ApS Copenhagen K1311 Denmark College of Arts Media and Design Northeastern University BostonMA02115 United States Department of Computer Science and Engineering New York University BrooklynNY11201 United States

This survey explores Procedural Content Generation via Machine Learning (PCGML), defined as the generation of game content using machine learning models trained on existing content. As the importance of PCG for game development increases, researchers explore new avenues for generating high-quality content with or without human involvement;this paper addresses the relatively new paradigm of using machine learning (in contrast with search-based, solver-based, and constructive methods). We focus on what is most often considered functional game content such as platformer levels, game maps, interactive fiction stories, and cards in collectible card games, as opposed to cosmetic content such as sprites and sound effects. In addition to using PCG for autonomous generation, co-creativity, mixed-initiative design, and compression, PCGML is suited for repair, critique, and content analysis because of its focus on modeling existing content. We discuss various data sources and representations that affect the generated content. Multiple PCGML methods are covered, including neural networks: long short-term memory (LSTM) networks, autoencoders, and deep convolutional networks;Markov models: n-grams and multi-dimensional Markov chains;clustering;and matrix factorization. Finally, we discuss open problems in PCGML, including learning from small datasets, lack of training data, multi-layered learning, style-transfer, parameter tuning, and PCG as a game mechanic. Copyright © 2017, The Authors. All rights reserved.

关键词： Knowledge representation

High Q impedance matching for RF energy harvesting applications

学校读者我要写书评

暂无评论

High Q impedance matching for RF energy harvesting applicati...

IEEE International Symposium on Wireless Systems (IDAACS-SWS)

作者： Christian Merz Gerald Kupris Faculty of Electrical Engineering Media Technology and Computer Science

RF energy harvesting is the extraction of small amounts of energy from an electromagnetic field. This energy can be used to power a tiny microsystem like a wireless and battery-less sensor. At the Deggendorf Institute of technology we have constructed RF energy harvesting circuits and tested their performance and features. The paper describes the building blocks of a RF energy harvesting system and discusses a possibility of improving the system. The method deals with high Q matching circuits in the 13.56 MHz band that increase the overall efficiency of RF energy harvesting circuits. Instead of common inductive and capacitive matching elements, high Q elements such like crystal resonators are used. A L matching circuit is analyzed mathematically in the paper and a RF energy harvesting network with high Q matching is simulated and manufactured and the advantages and disadvantages of high Q matching are discussed.

关键词： Resonant frequency Voltage measurement Energy harvesting Crystals Radio frequency Q-factor RLC circuits

Human and smart machine co-learning with brain computer interface

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Lee, Chang-Shing Wang, Mei-Hui Ko, Li-Wei Prof, Naoyuki Kubota Lin, Lu-An Kitaoka, Shinya Wang, Yu-Te Su, Shun-Feng Computer Science and Information Engineering National Cheng Kung University Tainan Taiwan Taiwan National Chiao Tung University Hsinchu Taiwan Institute of Bioinformatics and Systems Biology National Chiao Tung University Taiwan Nagoya University Nagoya Japan Department of System Design Tokyo Metropolitan University Tokyo Japan OASE Lab. NUTN Taiwan Dwango Media Village DWANGO Co. Ltd. Japan La Jolla Swartz Center for Computational Neuroscience UCSD United States Purdue University West LafayetteIN United States Department of Electrical Engineering National Taiwan University of Science and Technology Taiwan

Machine learning has become a very popular approach for cybernetics systems, and it has always been considered important research in the Computational Intelligence area. Nevertheless, when it comes to smart machines, it is not just about the methodologies. We need to consider systems and cybernetics as well as include human in the loop. The purpose of this article is as follows: (1) To integrate the open source Facebook AI Research (FAIR) DarkForest program of Facebook with Item Response Theory (IRT), to the new open learning system, namely, DDF learning system;(2) To integrate DDF Go with Robot namely Robotic DDF Go system;(3) To invite the professional Go players to attend the activity to play Go games on site with a smart machine. The research team will apply this technology to education, such as, playing games to enhance the children concentration on learning mathematics, languages, and other topics. With the detected brainwaves, the robot will be able to speak some words that are very much to the point for the students and to assist the teachers in classroom in the future. Copyright © 2018, The Authors. All rights reserved.

关键词： Social networking (online)

From zero-shot learning to conventional supervised classification: Unseen visual data synthesis

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Long, Yang Liu, Li Shao, Ling Shen, Fumin Ding, Guiguang Han, Jungong Department of Electronic and Electrical Engineering University of Sheffield School of Computing Sciences University of East Anglia Center for Future Media University of Electronic Science and Technology of China School of Software Tsinghua University Department of Computer Science and Digital Technologies Northumbria University

Robust object recognition systems usually rely on powerful feature extraction mechanisms from a large number of real images. However, in many realistic applications, collecting sufficient images for ever-growing new classes is unattainable. In this paper, we propose a new Zero-shot learning (ZSL) framework that can synthesise visual features for unseen classes without acquiring real images. Using the proposed Unseen Visual Data Synthesis (UVDS) algorithm, semantic attributes are effectively utilised as an intermediate clue to synthesise unseen visual features at the training stage. Hereafter, ZSL recognition is converted into the conventional supervised problem, i.e. the synthe-sised visual features can be straightforwardly fed to typical classifiers such as SVM. On four benchmark datasets, we demonstrate the benefit of using synthesised unseen data. Extensive experimental results suggest that our proposed approach significantly improve the state-of-the-art results. Copyright © 2017, The Authors. All rights reserved.

关键词： Object recognition

DIRECT INFERENCE ON COMPRESSIVE MEASUREMENTS USING CONVOLUTIONAL NEURAL NETWORKS

学校读者我要写书评

暂无评论

DIRECT INFERENCE ON COMPRESSIVE MEASUREMENTS USING CONVOLUTI...

IEEE International Conference on Image Processing

作者： Suhas Lohit Kuldeep Kulkarni Pavan Turaga School of Electrical Computer and Energy Engineering School of Arts Media and Engineering Arizona State University

ISBN: (纸本)9781467399623

Compressive imagers, e.g. the single-pixel camera (SPC), acquire measurements in the form of random projections of the scene instead of pixel intensities. Compressive Sensing (CS) theory allows accurate reconstruction of the image even from a small number of such projections. However, in practice, most reconstruction algorithms perform poorly at low measurement rates and are computationally very expensive. But perfect reconstruction is not the goal of high-level computer vision applications. Instead, we are interested in only determining certain properties of the image. Recent work has shown that effective inference is possible directly from the compressive measurements, without reconstruction, using correlational features. In this paper, we show that convolutional neural networks (CNNs) can be employed to extract discriminative non-linear features directly from CS measurements. Using these features, we demonstrate that effective high-level inference can be performed. Experimentally, using hand written digit recognition (MNIST dataset) and image recognition (ImageNet) as examples, we show that recognition is possible even at low measurement rates of about 0.1.

关键词： Compressive sensing Neural networks Compressed sensing Neural network Reconstruction algorithm perfect reconstruction images Recognition (Psychology)