Filters








2,827 Hits in 6.5 sec

Statistical Challenges in Modeling Big Brain Signals [article]

Zhaoxia Yu, Dustin Pluta, Tong Shen, Chuansheng Chen, Gui Xue, Hernando Ombao
2017 arXiv   pre-print
Brain signal data are inherently big: massive in amount, complex in structure, and high in dimensions. These characteristics impose great challenges for statistical inference and learning.  ...  In a recent work, Ting et al. (2017) proposed a method for high dimensional by finding low-dimensional representations via principal components analysis.  ...  By analyzing tensors, i.e., multi-array data, tensor regression can directly incorporate the inherent spatio-temporal structure of imaging data.  ... 
arXiv:1711.00432v1 fatcat:tozy4qe5dncdfnzb53lcooieke

TensorSplat: Spotting Latent Anomalies in Time

Danai Koutra, Evangelos E. Papalexakis, Christos Faloutsos
2012 2012 16th Panhellenic Conference on Informatics  
When we have multi-aspect data, e.g. who published which paper on which conference and on what year, how can we combine this information, in order to obtain good summaries thereof and unravel hidden anomalies  ...  "strange" behaviors.  ...  Tensor Applications Tensors are very powerful tools for the analysis of the continuously increasing (multi-dimensional) data that becomes available, and thus they are rather popular in the data mining  ... 
doi:10.1109/pci.2012.60 dblp:conf/pci/KoutraPF12 fatcat:ykpncrpf6fhkrlby5dfrp5xpfy

Tales of Two Cities: Using Social Media to Understand Idiosyncratic Lifestyles in Distinctive Metropolitan Areas

Tianran Hu, Eric Bigelow, Jiebo Luo, Henry Kautz
2017 IEEE Transactions on Big Data  
We discovered interesting human behavior patterns at both a larger scale and a finer granularity than is present in previous literature, some of which allow us to quantitatively compare the behaviors of  ...  In this paper, we examine and compare lifestyle behaviors of people living in cities of different sizes, utilizing freely available social media data as a large-scale, lowcost alternative to traditional  ...  Noulas et al. of [18] use Foursquare data to discover the behavioral habits of residents in London.  ... 
doi:10.1109/tbdata.2016.2580542 fatcat:tb7bat3uj5b7vjmfqtu7tv43ui

Tales of Two Cities: Using Social Media to Understand Idiosyncratic Lifestyles in Distinctive Metropolitan Areas [article]

Tianran Hu, Eric Bigelow, Jiebo Luo, Henry Kautz
2017 arXiv   pre-print
We discovered interesting human behavior patterns at both a larger scale and a finer granularity than is present in previous literature, some of which allow us to quantitatively compare the behaviors of  ...  In this paper, we examine and compare lifestyle behaviors of people living in cities of different sizes, utilizing freely available social media data as a large-scale, low-cost alternative to traditional  ...  Noulas et al. of [18] use Foursquare data to discover the behavioral habits of residents in London.  ... 
arXiv:1701.06236v1 fatcat:vfddxhxqvfddbhme6x5ey5pzt4

Non-Negative Tensor Factorization for Human Behavioral Pattern Mining in Online Games

2018 Information  
Here, we leverage the Non-negative Tensor Factorization to detect hidden correlated behaviors of playing in a well-known game: League of Legends.  ...  A crucial problem is the extraction of activity patterns that characterize this type of data, in an interpretable way.  ...  Analogously to matrix factorization, tensor factorization allows to approximate data in a lower dimensional space.  ... 
doi:10.3390/info9030066 fatcat:msy5xcgxvnfmthzvlgwmckgqzi

On Overfitting and Asymptotic Bias in Batch Reinforcement Learning with Partial Observability (Extended Abstract)

Vincent Francois-Lavet, Guillaume Rabusseau, Joelle Pineau, Damien Ernst, Raphael Fonteneau
2020 Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence  
) and a term due to overfitting (additional suboptimality due to limited data).  ...  In the context of reinforcement learning with partial observability, this paper provides an analysis of the tradeoff between these two error sources.  ...  The latter one directly represents sequential response data as a high-order tensor, and factorizes tensors to obtain latent embeddings for users and items.  ... 
doi:10.24963/ijcai.2020/695 dblp:conf/ijcai/0001Z20 fatcat:yx2wihhuobgmjjh4aevkbr33g4

Non-negative Tensor Factorization for Human Behavioral Pattern Mining in Online Games [article]

Anna Sapienza, Alessandro Bessi, Emilio Ferrara
2017 arXiv   pre-print
Here, we propose to exploit tensor decomposition techniques, and in particular Non-negative Tensor Factorization, to discover hidden correlated behavioral patterns of play in a popular game: League of  ...  A major problem is extracting meaningful patterns of activity from this type of data, in a way that is also easy to interpret.  ...  Non-negative tensor factorization allows to identify correlation in the data at di erent levels [14] : on the one hand, the application of the NTF helps in the identi cation of hidden topological structures  ... 
arXiv:1702.05695v1 fatcat:o5oe3udvgzgidcikqruwmjisfq

AI and Deep Learning for Urban Computing [chapter]

Senzhang Wang, Jiannong Cao
2021 The Urban Book Series  
AbstractIn the big data era, with the large volume of available data collected by various sensors deployed in urban areas and the recent advances in AI techniques, urban computing has become increasingly  ...  Then we briefly introduce the AI techniques that are widely used in urban computing, including supervised learning, semi-supervised learning, unsupervised learning, matrix factorization, graphic models  ...  The value hidden in the data can be low and may require carefully designed machine-learning or data-mining methods to discover useful knowledge from the massive data.  ... 
doi:10.1007/978-981-15-8983-6_43 fatcat:uq7j3hvsvzfl5lq33omx64un3i

Manifold Modeling in Embedded Space: A Perspective for Interpreting Deep Image Prior [article]

Tatsuya Yokota, Hidekata Hontani, Qibin Zhao, Andrzej Cichocki
2020 arXiv   pre-print
In spite of its simplicity, the image/tensor completion, super-resolution, deconvolution, and denoising results of MMES are quite similar even competitive to DIP in our extensive experiments, and these  ...  Deep image prior (DIP), which utilizes a deep convolutional network (ConvNet) structure itself as an image prior, has attracted attentions in computer vision and machine learning communities.  ...  Manifold modeling for this task is based on the idea that the dimensionality of many data sets is only artificially high.  ... 
arXiv:1908.02995v2 fatcat:nfv6i5xqtre3berkpoj2jmyedi

Tensor-based anomaly detection: An interdisciplinary survey

Hadi Fanaee-T, João Gama
2016 Knowledge-Based Systems  
However, if data includes tensor (multiway) structure (e.g. space-time-measurements), some meaningful anomalies may remain invisible with these methods.  ...  Although tensor-based anomaly detection (TAD) has been applied within a variety of disciplines over the last twenty years, it is not yet recognized as a formal category in anomaly detection.  ...  Complex and multiway structure of hu-man behaviors was probably the first motivation for use of tensors in data analysis.  ... 
doi:10.1016/j.knosys.2016.01.027 fatcat:lejxxae63jcutfx2ncahownt7e

Understanding Urban Dynamics via Context-Aware Tensor Factorization with Neighboring Regularization

Jingyuan Wang, Junjie Wu, Ze Wang, Fei Gao, Zhang Xiong
2019 IEEE Transactions on Knowledge and Data Engineering  
In this paper, we propose a Neighbor-Regularized and context-aware Non-negative Tensor Factorization model (NR-cNTF) to discover interpretable urban dynamics from urban heterogeneous data.  ...  This is enabled by high-quality Tucker factorizations regularized by both POI-based urban contexts and geographically neighboring relations.  ...  The focal point here is to use tensor factorization to discover latent lower-dimensional factors from higher-dimensional multi-aspect data sets.  ... 
doi:10.1109/tkde.2019.2915231 fatcat:ghoaa73yorbf3gkpkmutdlym3y

General-Purpose Unsupervised Cyber Anomaly Detection via Non-Negative Tensor Factorization

Maksim E. Eren, Juston S. Moore, Erik Skau, Elisabeth Moore, Manish Bhattarai, Gopinath Chennupati, Boian S. Alexandrov
2022 Digital Threats: Research and Practice  
Non-negative tensor factorization, on the other hand, is a powerful unsupervised machine learning method that naturally models multi-dimensional data, capturing complex and multi-faceted details of behavior  ...  However, approaches proposed to date based on probabilistic matrix factorization are limited by the information conveyed in a two-dimensional space.  ...  Tensor factorization decomposes high-dimensional data into lower-dimensional components (usually 2D factor matrices), where the factor matrices carry the latent features in each tensor dimension.  ... 
doi:10.1145/3519602 fatcat:uxt4l4g4pzayjhvxqmmgfksfpy

Adaptive Granularity in Tensors: A Quest for Interpretable Structure [article]

Ravdeep Pasricha, Ekta Gujral, Evangelos E. Papalexakis
2022 arXiv   pre-print
in high-quality tensors.  ...  In all the cases, our proposed method constructs tensors that have very high structure quality.  ...  intrinsic hidden structure of the data, and 2) extrinsic measures, where the quality is measured by how well the computed decomposition factors perform in a predictive task.  ... 
arXiv:1912.09009v2 fatcat:oow6xgwjvfe3za45lsgyaquboi

Understanding Multilingual Social Networks in Online Immigrant Communities

Evangelos Papalexakis, A. Seza Doğruöz
2015 Proceedings of the 24th International Conference on World Wide Web - WWW '15 Companion  
Using a novel method (tensor analysis), we reveal the social network structure of an online multilingual discussion forum which hosts an immigrant community in the Netherlands.  ...  In addition to the network structure, we automatically discover and categorize monolingual and bilingual sub-communities and track their formation, evolution and dissolution over a long period of time.  ...  As expected, the data we are dealing with are high dimensional and highly sparse.  ... 
doi:10.1145/2740908.2743004 dblp:conf/www/PapalexakisD15 fatcat:j3ilssq53beuno4siesa4j5nre

Guaranteed Scalable Learning of Latent Tree Models [article]

Furong Huang, Niranjan U.N., Ioakeim Perros, Robert Chen, Jimeng Sun, Anima Anandkumar
2019 arXiv   pre-print
We present an integrated approach for structure and parameter estimation in latent tree graphical models.  ...  Our bulk asynchronous parallel algorithm is implemented in parallel and the parallel computation complexity increases only logarithmically with the number of variables and linearly with dimensionality  ...  Anandkumar is supported in part by Bren endowed chair, Darpa PAI, Raytheon, and Microsoft, Google and Adobe faculty fellowships.  ... 
arXiv:1406.4566v4 fatcat:lupn2z2d7fagfnkvgnhjz4vqg4
« Previous Showing results 1 — 15 out of 2,827 results