169 Hits in 10.4 sec

Complete the Missing Half: Augmenting Aggregation Filtering with Diversification for Graph Convolutional Networks [article]

Sitao Luan, Mingde Zhao, Chenqing Hua, Xiao-Wen Chang, Doina Precup
2020 arXiv   pre-print
The core operation of current Graph Neural Networks (GNNs) is the aggregation enabled by the graph Laplacian or message passing, which filters the neighborhood node information.  ...  Such augmentation replaces the aggregation with a two-channel filtering process that, in theory, is beneficial for enriching the node representations.  ...  FB-GNN framework can easily be plugged into spatial methods, with LP filter for aggregation operation and HP filter for diversification operation.  ... 
arXiv:2008.08844v3 fatcat:ehtmid3prvcdtlxemqeyz7vp2q

MARVEL - D3.1: Multimodal and privacy-aware audio-visual intelligence – initial version

Alexandros Iosifidis
2022 Zenodo  
as methodologies for improving the training and efficiency of AI models under supervised, unsupervised, and cross-modal contrastive learning settings.  ...  This document describes the initial version of the methodologies pro- posed by MARVEL partners towards the realisation of the Audio, Visual and Multimodal AI Subsystem of the MARVEL architecture.  ...  Such scalability features allow for extreme model compression and optimisation, while decoupling parameter count and computational cost in alignment with the harware-aware scaling paradigm.  ... 
doi:10.5281/zenodo.6821317 fatcat:eia7rkk5lfbg7khs3qcat5qd3m

A Roadmap for Big Model [article]

Sha Yuan, Hanyu Zhao, Shuai Zhao, Jiahong Leng, Yangxiao Liang, Xiaozhi Wang, Jifan Yu, Xin Lv, Zhou Shao, Jiaao He, Yankai Lin, Xu Han (+88 others)
2022 arXiv   pre-print
With the rapid development of deep learning, training Big Models (BMs) for multiple downstream tasks becomes a popular paradigm.  ...  In this paper, we cover not only the BM technologies themselves but also the prerequisites for BM training and applications with BMs, dividing the BM review into four parts: Resource, Models, Key Technologies  ...  Therefore, it is important to automatically complete the missing knowledge in the knowledge graphs.  ... 
arXiv:2203.14101v4 fatcat:rdikzudoezak5b36cf6hhne5u4

Detecting Abnormal Social Network Accounts with Hurst of Interest Distribution

Xiujuan Wang, Yi Sui, Yuanrui Tao, Qianqian Zhang, Jianhua Wei, Manjit Kaur
2021 Security and Communication Networks  
With the rapid development of the Internet since the beginning of the 21st century, social networks have provided a significant amount of convenience for work, study, and entertainment.  ...  Detecting abnormal accounts on social networks in a timely manner can effectively prevent the occurrence of malicious Internet events.  ...  With the rise of deep learning, a large number of researchers have considered using deep-learning models to automatically model graph data, including graph embedding [21] and graph neural networks [  ... 
doi:10.1155/2021/6653430 fatcat:7cr6sdbel5gqfmlbjvct5bze5y

MMDF2018 Workshop Report [article]

Chun-An Chou, Xiaoning Jin, Amy Mueller, Sarah Ostadabbas
2018 arXiv   pre-print
Driven by the recent advances in smart, miniaturized, and mass produced sensors, networked systems, and high-speed data communication and computing, the ability to collect and process larger volumes of  ...  ., the development of generalized solutions) can only be achieved via a high level cross-disciplinary aggregation of learnings, and this workshop was proposed at an opportune time as many domains have  ...  Kramer), and College of Engineering (and the Dean of Engineering, Professor Nadine Aubry) for providing invaluable support with funds, organization, and logistics.  ... 
arXiv:1808.10721v1 fatcat:w4pmt5vqgvcn7nltkisitjpyam

An Introduction to Sensor Data Analytics [chapter]

Charu C. Aggarwal
2012 Managing and Mining Sensor Data  
Primitive sensor events need to be filtered, aggregated and correlated to generate more semantically rich complex events to facilitate the requirements of up-streaming applications.  ...  Furthermore, MIST proposes an in-network index structure for indexing the HMMs. This index can be used for improving the performance of query processing.  ...  The U.S. Government is authorized to reproduce and distribute reprints for Government purposes notwithstanding any copyright notation here on.  ... 
doi:10.1007/978-1-4614-6309-2_1 fatcat:pfbx566yfzgqpnjcuzonmxr23q

Multi-Agent Autonomy: Advancements and Challenges in Subterranean Exploration [article]

Michael T. Ohradzansky, Eugene R. Rush, Danny G. Riley, Andrew B. Mills, Shakeeb Ahmad, Steve McGuire, Harel Biggie, Kyle Harlow, Michael J. Miles, Eric W. Frew, Christoffer Heckman, J. Sean Humbert
2021 arXiv   pre-print
The team presents two navigation algorithms in the form of a metric-topological graph-based planner and a continuous frontier-based planner.  ...  DARPA is seeking to change that with the Subterranean Challenge, by providing roboticists the opportunity to support civilian and military first responders in complex and high-risk underground scenarios  ...  A special thanks to Daniel Torres, Cesar Galant, Zoe Turin, for assisting with the design, testing, and deployments of our platforms and algorithms.  ... 
arXiv:2110.04390v1 fatcat:53a4flk2jrhbdkhmxfvipf4m2a

Commercial Visual Analytics Systems-Advances in the Big Data Analytics Field

Michael Behrisch, Dirk Streeb, Florian Stoffel, Daniel Seebacher, Brian Matejek, Stefan Hagen Weber, Sebastian Mittelstaedt, Hanspeter Pfister, Daniel Keim
2018 IEEE Transactions on Visualization and Computer Graphics  
We also investigate previously unavailable products to paint a more complete picture of the commercial VA landscape.  ...  We evaluate new product versions on established evaluation criteria, such as available features, performance, and usability, to extend on and assure comparability with the previous survey.  ...  These deep learning approaches, such as described in the form of convolutional neural networks (CNNs) for image classification [55] , [56] or in the form of recurrent neural networks (RNNs) for text  ... 
doi:10.1109/tvcg.2018.2859973 pmid:30059307 fatcat:w22e7ha7i5cxvokajils6mmbzq

Automatic Identification of Harmful, Aggressive, Abusive, and Offensive Language on the Web: A Survey of Technical Biases Informed by Psychology Literature

Agathe Balayn, Jie Yang, Zoltan Szlavik, Alessandro Bozzon
2021 ACM Transactions on Social Computing  
resulting technical biases in the design of machine learning classification models and the dataset created for their training.  ...  Finally, we discuss diverse research opportunities for the computer science community and reflect on broader technical and structural issues.  ...  the network graph) could be further explored by adopting the methodology followed in psychology.  ... 
doi:10.1145/3479158 fatcat:oio6v4ifrravxk2abh4sa35tzq

Cryptocurrency trading: a comprehensive survey

Fan Fang, Carmine Ventre, Michail Basios, Leslie Kanthan, David Martinez-Rego, Fan Wu, Lingbo Li
2022 Financial Innovation  
Although they have some commonalities with more traditional assets, they have their own separate nature and their behaviour as an asset is still in the process of being understood.  ...  AbstractIn recent years, the tendency of the number of financial institutions to include cryptocurrencies in their portfolios has accelerated.  ...  The RNN with ten hidden layers is optimised for the setting and the neural network augmented by VAR allows the network to be shallower, quicker and to have a better prediction than an RNN.  ... 
doi:10.1186/s40854-021-00321-6 fatcat:d3d2pkxy5fgcfa4s6gi4h2snua

Cryptocurrency Trading: A Comprehensive Survey [article]

Fan Fang, Carmine Ventre, Michail Basios, Leslie Kanthan, Lingbo Li, David Martinez-Regoband, Fan Wu
2022 arXiv   pre-print
Although they have some commonalities with more traditional assets, they have their own separate nature and their behaviour as an asset is still in the process of being understood.  ...  In recent years, the tendency of the number of financial institutions including cryptocurrencies in their portfolios has accelerated.  ...  The survey represents a quick way to get familiar with the literature on cryptocurrency trading and can motivate more researchers to contribute to the pressing problems in the area, for example along the  ... 
arXiv:2003.11352v5 fatcat:l7eih2yoazbq5i5lv4wh7c24ha

Enabling Technologies for Ultra-Reliable and Low Latency Communications: From PHY and MAC Layer Perspectives

Gordon J. Sutton, Jie Zeng, Ren Ping Liu, Wei Ni, Diep N. Nguyen, Beeshanga A. Jayawickrama, Xiaojing Huang, Mehran Abolhasan, Zhang Zhang, Eryk Dutkiewicz, Tiejun Lv
2019 IEEE Communications Surveys and Tutorials  
The paper evaluates the relevant PHY and MAC techniques for their ability to improve the reliability and reduce the latency.  ...  Future 5th generation (5G) networks are expected to enable three key services -enhanced mobile broadband (eMBB), massive machine type communications (mMTC) and ultra-reliable and low latency communications  ...  An ultra-dense network is simulated, showing that NOMA with full-duplex can have much higher sum rate than both OMA with halfduplex and NOMA with half-duplex.  ... 
doi:10.1109/comst.2019.2897800 fatcat:jgaahgi27rev3ieyrisn2yvh5m

Computational Socioeconomics [article]

Jian Gao, Yi-Cheng Zhang, Tao Zhou
2019 arXiv   pre-print
Uncovering the structure of socioeconomic systems and timely estimation of socioeconomic status are significant for economic development.  ...  This review, together with pioneering works we have highlighted, will draw increasing interdisciplinary attentions and induce a methodological shift in future socioeconomic studies.  ...  [599] developed the convolutional network (ConvNet) architecture to transform MP data into high-level features for each week and then aggregated patterns across weeks by reusing the same convolutional  ... 
arXiv:1905.06166v1 fatcat:kvhy2hpzgvg2vnqhdjfyjfidqi

Machine Learning for Physical Layer in 5G and beyond Wireless Networks: A Survey

Jawad Tanveer, Amir Haider, Rashid Ali, Ajung Kim
2021 Electronics  
With their enhanced speed, 5G networks are prone to various research challenges.  ...  Fifth-generation networks potentially merge multiple networks on a single platform, providing a landscape for seamless connectivity, particularly for high-mobility devices.  ...  [200] proposed a DL model (CNN-DMA) to detect malware attacks based on a classifier-the Convolution Neural Network (CNN)-and [201] proposed a multi-scale convolutional neural network framework for  ... 
doi:10.3390/electronics11010121 fatcat:ho7yyefbm5gf3bz3t2h3cxqtjq

A Comprehensive Survey on Deep Music Generation: Multi-level Representations, Algorithms, Evaluations, and Future Directions [article]

Shulei Ji, Jing Luo, Xinyu Yang
2020 arXiv   pre-print
Previous surveys have explored the network models employed in the field of automatic music generation.  ...  In addition, we summarize the datasets suitable for diverse tasks, discuss the music representations, the evaluation methods as well as the challenges under different levels, and finally point out several  ...  Music Inpainting/Completion Music inpainting/completion refers to filling the missing information in a piece of music, which is more in line with the human creation process.  ... 
arXiv:2011.06801v1 fatcat:cixou3d2jzertlcpb7kb5x5ery
« Previous Showing results 1 — 15 out of 169 results