检索结果-内蒙古大学图书馆

Improving individual predictions using social networks assortativity

学校读者我要写书评

暂无评论

Improving individual predictions using social networks assor...

International Workshop on Self-Organizing Maps and Learning Vector Quantization, Clustering and data Visualization (WSOM)

作者： Dounia Mulders Cyril de Bodt Johannes Bjelland Alex Sandy Pentland Michel Verleysen Yves-Alexandre de Montjoye ICTEAM institute Université catholiaue de Louvain Telenor Research MIT Media Lab Massachusetts Institute of Technology Department of Computing Data Science Institute

ISBN: (纸本)9781509066391

Social networks are known to be assortative with respect to many attributes, such as age, weight, wealth, level of education, ethnicity and gender. This can be explained by influences and homophilies. Independently of its origin, this assortativity gives us information about each node given its neighbors. Assortativity can thus be used to improve individual predictions in a broad range of situations, when data are missing or inaccurate. This paper presents a general framework based on probabilistic graphical models to exploit social network structures for improving individual predictions of node attributes. Using this framework, we quantify the assortativity range leading to an accuracy gain in several situations. We finally show how specific characteristics of the network can improve performances further. For instance, the gender assortativity in real-world mobile phone data changes significantly according to some communication attributes. In this case, individual predictions with 75% accuracy are improved by up to 3%.

关键词： Mobile handsets Graphical models Tuning Probabilistic logic Twitter Predictive models

Investigating different syntactic context types and context representations for learning word embeddings

学校读者我要写书评

暂无评论

Investigating different syntactic context types and context ...

2017 Conference on Empirical Methods in Natural Language Processing, EMNLP 2017

作者： Li, Bofang Liu, Tao Zhao, Zhe Tang, Buzhou Drozd, Aleksandr Rogers, Anna Du, Xiaoyong School of Information Renmin University of China China Key Laboratory of Data Engineering and Knowledge Engineering MOE China Shenzhen Graduate School Harbin Institute of Technology China Global Scientific Information and Computing Center Tokyo Institute of Technology China Department of Computer Science University of Massachusetts Lowell United States

ISBN: (纸本)9781945626838

The number of word embedding models is growing every year. Most of them are based on the co-occurrence information of words and their contexts. However, it is still an open question what is the best definition of context. We provide a systematical investigation of 4 different syntactic context types and context representations for learning word embeddings. Comprehensive experiments are conducted to evaluate their effectiveness on 6 extrinsic and intrinsic tasks. We hope that this paper, along with the published code, would be helpful for choosing the best context type and representation for a given task. © 2017 Association for Computational Linguistics.

关键词： Embeddings

Low-resolution face recognition in the wild via selective knowledge distillation

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Ge, Shiming Zhao, Shengwei Li, Chenyu Li, Jia Institute of Information Engineering Chinese Academy of Sciences Beijing100095 China Institute of Information Engineering Chinese Academy of Sciences School of Cyber Security at University of Chinese Academy of Sciences China State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beihang University Beijing Advanced Innovation Center for Big Data and Brain Computing Beihang University Beijing100191 China

Typically, the deployment of face recognition models in the wild needs to identify low-resolution faces with extremely low computational cost. To address this problem, a feasible solution is compressing a complex face model to achieve higher speed and lower memory at the cost of minimal performance drop. Inspired by that, this paper proposes a learning approach to recognize low-resolution faces via selective knowledge distillation. In this approach, a two-stream convolutional neural network (CNN) is first initialized to recognize high-resolution faces and resolution-degraded faces with a teacher stream and a student stream, respectively. The teacher stream is represented by a complex CNN for high-accuracy recognition, and the student stream is represented by a much simpler CNN for low-complexity recognition. To avoid significant performance drop at the student stream, we then selectively distil the most informative facial features from the teacher stream by solving a sparse graph optimization problem, which are then used to regularize the fine-tuning process of the student stream. In this way, the student stream is actually trained by simultaneously handling two tasks with limited computational resources: approximating the most informative facial cues via feature regression, and recovering the missing facial cues via low-resolution face classification. Experimental results show that the student stream performs impressively in recognizing low-resolution faces and costs only 0.15MB memory and runs at 418 faces per second on CPU and 9, 433 faces per second on GPU. Copyright © 2018, The Authors. All rights reserved.

关键词： Students

Snowmass 2021 Computational Frontier CompF03 Topical Group Report: Machine Learning

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Shanahan, Phiala Terao, Kazuhiro Whiteson, Daniel Aarts, Gert Adelmann, Andreas Akchurin, N. Alexandru, Andrei Amram, Oz Andreassen, Anders Apresyan, Artur Avestruz, Camille Bartoldus, Rainer Bechtol, Keith Benkendorfer, Kees Benelli, Gabriele Bernius, Catrin Bogatskiy, Alexander Bortolato, Blaz Boyda, Denis Brooijmans, Gustaaf Calafiura, Paolo Calì, Salvatore Canelli, Florencia Chachamis, Grigorios Chekanov, S.V. Chen, Deming Chen, Thomas Y. Ćiprijanović, Aleksandra Collins, Jack H. Connolly, Andrew J. Coughlin, Michael Dai, Biwei Damgov, J. DeZoort, Gage Diaz, Daniel Dillon, Barry M. Dinu, Ioan-Mihail Dong, Zhongtian Donini, Julien Duarte, Javier Dugad, S. Dvorkin, Cora Faroughy, D.A. Feickert, Matthew Feng, Yongbin Fenton, Michael Foreman, Sam De Freitas, Felipe F. Funcke, Lena Gc, P. Gandrakota, Abhijith Ganguly, Sanmay Garrison, Lehman H. Gessner, Spencer Ghosh, Aishik Gonsk, Julia Graham, Matthew Gray, Lindsey Grönroos, S. Hackett, Daniel C. Harris, Philip Hauck, Scott Herwig, Christian Holzman, Burt Hopkins, Walter Hsu, Shih-Chieh Huang, Jin Huang, Yi Jin, Xiao-Yong Kagan, Michael Kah, Alan Kamenik, Jernej F. Kansal, Raghav Karagiorgi, Georgia Kasieczka, Gregor Katsavounidis, Erik Khoda, Elham E. Khosa, Charanjit K. Kipf, Thomas Komiske, Patrick Komm, Matthias Kondor, Risi Kourlitis, Evangelos Krause, Claudius Lamichhane, K. Le Pottier, Luc Lin, Meifeng Lin, Yin Liu, Mia Lu, Nan Lucini, Biagio Martinez, J. Martín-Ramiro, Pablo Matevc, Andrej McCormack, William Patrick Metodiev, Eric Mikuni, Vinicius Miller, David W. Mishra-Sharma, Siddharth Mukherjee, Samadrita Murnane, Daniel Nachman, Benjamin Narayan, Gautham Neubauer, Mark Ngadiuba, Jennifer Norberg, Scarlet Nord, Brian Ochoa, Inês Offermann, Jan T. Park, Sang Eon Pedro, Kevin Peña, Cristían Perloff, Alexx Pettee, Mariel Pierini, Maurizio Quast, T. Rankin, Dylan Ren, Yihui Rieger, Marcel Vlimant, Jean-Roch Roy, Avik Sanz, Veronica Sarda, Nilai Savard, Claire Scheinker, Alexander Seljak, Uroš Sheldon, Brian Shih, David Shimmin, Chase Smolkovic, Aleks Ste Swansea University SwanseaSA2 8PP United Kingdom Villazzano38123 Italy Paul Scherrer Institute PSI Villigen5232 Switzerland Texas Tech University LubbockTX79409 United States The George Washington University WashingtonDC20052 United States University of Maryland College ParkMD20742 United States The Johns Hopkins University BaltimoreMD21211 United States Google Mountain ViewCA94043 United States Fermi National Accelerator Laboratory BataviaIL60510 United States University of Michigan Ann ArborMI48109 United States SLAC National Accelerator Laboratory StanfordCA94309 United States University of Wisconsin-Madison 1150 University Avenue MadisonWI53706-1390 United States Lawrence Berkeley National Laboratory BerkeleyCA94720 United States Reed College PortlandOR97202 United States Flatiron Institute 162 5th Avenue New YorkNY10010 United States Jožef Stefan Institute Jamova 39 Ljubljana1000 Slovenia Argonne National Laboratory ArgonneIL60439 United States The NSF AI Institute for Artificial Intelligence and Fundamental Interactions United States Nevis Laboratories Columbia University 136 S Broadway IrvingtonNY United States Massachusetts Institute of Technology 77 Massachusetts Ave CambridgeMA02139 United States University of Zurich Winterthurerstrasse 190 Zurich8057 Switzerland Switzerland University of Illinois at Urbana-Champaign ChampaignIL61820 United States University of Washington SeattleWA98195 United States University of Minnesota MinneapolisMN55455 United States Berkeley Center for Cosmological Physics University of California Berkeley United States Princeton University PrincetonNJ08544 United States University of California La Jolla San DiegoCA92093 United States University of Heidelberg Heidelberg Germany University of Kansas 1251 Wescoe Hall Dr. LawrenceKS66045 United States Université Clermont Auvergne France Tata Institute of Fundamental Research Mumbai400005 India Harvard University

The rapidly-developing intersection of machine learning (ML) with high-energy physics (HEP) presents both opportunities and challenges to our community. Far beyond applications of standard ML tools to HEP problems, genuinely new and potentially revolutionary approaches are being developed by a generation of talent literate in both fields. There is an urgent need to support the needs of the interdisciplinary community driving these developments, including funding dedicated research at the intersection of the two fields, investing in high-performance computing at universities and tailoring allocation policies to support this work, developing of community tools and standards, and providing education and career paths for young researchers attracted by the intellectual vitality of machine learning for high energy physics. © 2022, CC BY.

关键词： High energy physics

HoloScope: Topology-and-Spike Aware Fraud Detection

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Liu, Shenghua Hooi, Bryan Faloutsos, Christos CAS Key Laboratory of Network Data Science & Technology Institute of Computing Technology Chinese Academy of Sciences Carnegie Mellon University

As online fraudsters invest more resources, including purchasing large pools of fake user accounts and dedicated IPs, fraudulent attacks become less obvious and their detection becomes increasingly challenging. Existing approaches such as average degree maximization suffer from the bias of including more nodes than necessary, resulting in lower accuracy and increased need for manual verification. Hence, we propose HoloScope, which uses information from graph topology and temporal spikes to more accurately detect groups of fraudulent users. In terms of graph topology, we introduce "contrast suspiciousness," a dynamic weighting approach, which allows us to more accurately detect fraudulent blocks, particularly low-density blocks. In terms of temporal spikes, HoloScope takes into account the sudden bursts and drops of fraudsters’ attacking patterns. In addition, we provide theoretical bounds for how much this increases the time cost needed for fraudsters to conduct adversarial attacks. Additionally, from the perspective of ratings, HoloScope incorporates the deviation of rating scores in order to catch fraudsters more accurately. Moreover, HoloScope has a concise framework and sub-quadratic time complexity, making the algorithm reproducible and scalable. Extensive experiments showed that HoloScope achieved significant accuracy improvements on synthetic and real data, compared with state-of-the-art fraud detection methods. Copyright © 2017, The Authors. All rights reserved.

关键词： Drops

Marked temporal dynamics modeling based on recurrent neural network

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Wang, Yongqing Liu, Shenghua Shen, Huawei Cheng, Xueqi CAS Key Laboratory of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences Beijing100190 China

We are now witnessing the increasing availability of event stream data, i.e., a sequence of events with each event typically being denoted by the time it occurs and its mark information (e.g., event type). A fundamental problem is to model and predict such kind of marked temporal dynamics, i.e., when the next event will take place and what its mark will be. Existing methods either predict only the mark or the time of the next event, or predict both of them, yet separately. Indeed, in marked temporal dynamics, the time and the mark of the next event are highly dependent on each other, requiring a method that could simultaneously predict both of them. To tackle this problem, in this paper, we propose to model marked temporal dynamics by using a mark-specific intensity function to explicitly capture the dependency between the mark and the time of the next event. Extensive experiments on two datasets demonstrate that the proposed method outperforms state-of-the-art methods at predicting marked temporal dynamics. Copyright © 2017, The Authors. All rights reserved.

关键词： Forecasting

Distinguishing unitary gates on the IBM quantum processor

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Liu, Shusen Li, Yinan Duan, Runyao School of Data and Computer Science Sun Yat-sen University Guangzhou Guangdong510006 China Centre for Quantum Software and Information Faculty of Engineering and Information Technology University of Technology SydneyNSW2007 Australia Centrum Wiskunde & Informatica and Research Center for Quantum Software Netherlands Institute for Quantum Computing Baidu Inc. Beijing100193 China

An unknown unitary gates, which is secretly chosen from several known ones, can always be distinguished perfectly. In this paper, we implement such a task on IBM's quantum processor. More precisely, we experimentally demonstrate the discrimination of two qubit unitary gates, the identity gate and the 2/3π-phase shift gate, using two discrimination schemes - the parallel scheme and the sequential scheme. We program these two schemes on the ibmqx4, a 5-qubit superconducting quantum processor via IBM cloud, with the help of the QSI modules [S. Liu et al., arXiv:1710.09500, 2017]. We report that both discrimination schemes achieve success probabilities at least 85%. Copyright © 2018, The Authors. All rights reserved.

关键词： Program processors

EagleMine: Vision-guided mining in large graphs

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Feng, Wenjie Liu, Shenghua Faloutsos, Christos Hooi, Bryan Shen, Huawei Cheng, Xueqi Cas Key Laboratory of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences School of Computer Science Carnegie Mellon University

Given a graph with millions of nodes, what patterns exist in the distributions of node characteristics, and how can we detect them and separate anomalous nodes in a way similar to human vision? In this paper, we propose a vision-guided algorithm, EagleMine, to summarize micro-cluster patterns in two-dimensional histogram plots constructed from node features in a large graph. EagleMine utilizes a water-level tree to capture cluster structures according to vision-based intuition at multi-resolutions. EagleMine traverses the waterlevel tree from the root and adopts statistical hypothesis tests to determine the optimal clusters that should be fitted along the path, and summarizes each cluster with a truncated Gaussian distribution. Experiments on real data show that our method can find truncated and overlapped elliptical clusters, even when some baseline methods split one visual cluster into pieces with Gaussian spheres. To identify potentially anomalous micro-clusters, EagleMine also a designates score to measure the suspiciousness of outlier groups (i.e. node clusters) and outlier nodes, detecting bots and anomalous users with high accuracy in the real Microblog data. Copyright © 2017, The Authors. All rights reserved.

关键词： Statistics

Not just privacy: Improving performance of private deep learning in mobile cloud

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Wang, Ji Zhang, Jianguo Bao, Weidong Zhu, Xiaomin Cao, Bokai Yu, Philip S. College of Systems Engineering National University of Defense Technology Changsha China Department of Computer Science University of Illinois at Chicago Chicago United States College of Systems Engineering State Key Laboratory of High Performance Computing National University of Defense Technology Changsha China Facebook Inc. Menlo Park United States Department of Computer Science University of Illinois at Chicago Chicago United States Institute for Data Science Tsinghua University Beijing China

The increasing demand for on-device deep learning services calls for a highly efficient manner to deploy deep neural networks (DNNs) on mobile devices with limited capacity. The cloud-based solution is a promising approach to enabling deep learning applications on mobile devices where the large portions of a DNN are offloaded to the cloud. However, revealing data to the cloud leads to potential privacy risk. To benefit from the cloud data center without the privacy risk, we design, evaluate, and implement a cloud-based framework Arden which partitions the DNN across mobile devices and cloud data centers. A simple data transformation is performed on the mobile device, while the resource-hungry training and the complex inference rely on the cloud data center. To protect the sensitive information, a lightweight privacy-preserving mechanism consisting of arbitrary data nullification and random noise addition is introduced, which provides strong privacy guarantee. A rigorous privacy budget analysis is given. Nonetheless, the private perturbation to the original data inevitably has a negative impact on the performance of further inference on the cloud side. To mitigate this influence, we propose a noisy training method to enhance the cloud-side network robustness to perturbed data. Through the sophisticated design, Arden can not only preserve privacy but also improve the inference performance. To validate the proposed Arden, a series of experiments based on three image datasets and a real mobile application are conducted. The experimental results demonstrate the effectiveness of Arden. Finally, we implement Arden on a demo system to verify its practicality. Copyright © 2018, The Authors. All rights reserved.

关键词： Deep neural networks