Filters








822 Hits in 2.2 sec

Learning Representations from Imperfect Time Series Data via Tensor Rank Regularization [article]

Paul Pu Liang, Zhun Liu, Yao-Hung Hubert Tsai, Qibin Zhao, Ruslan Salakhutdinov, Louis-Philippe Morency
2019 arXiv   pre-print
We design a model to learn such tensor representations and effectively regularize their rank.  ...  Our method is based on the observation that high-dimensional multimodal time series data often exhibit correlations across time and modalities which leads to low-rank tensor representations.  ...  In this paper, we propose a model called the Temporal Tensor Fusion Network (T2FN) that builds tensor representations from multimodal time series data.  ... 
arXiv:1907.01011v1 fatcat:ifmw2tzvbnbmrocj54ioybkiia

Table of Contents [EDICS]

2020 IEEE Transactions on Signal Processing  
Brie 931 Learning Nonnegative Factors From Tensor Data: Probabilistic Modeling and Inference Algorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  ...  C. de Lamare 81 Tensor Completion From Regular Sub-Nyquist Samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  ...  Ren Other Areas and Applications Block-Randomized Stochastic Proximal Gradient for Low-Rank Tensor Factorization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  ... 
doi:10.1109/tsp.2020.3045363 fatcat:wcnvdcy3rvhblh7rtxfe6gz4re

2020 Index IEEE Transactions on Signal Processing Vol. 68

2020 IEEE Transactions on Signal Processing  
., One-Step Prediction for Discrete Time-Varying Nonlinear Systems With Unknown Inputs and Correlated Noises; TSP  ...  ., +, TSP 2020 4268-4282 Tensor Completion From Regular Sub-Nyquist Samples.  ...  ., +, TSP 2020 1-16 Image representation A Low-Rank Tensor Dictionary Learning Method for Hyperspectral Image Denoising.  ... 
doi:10.1109/tsp.2021.3055469 fatcat:6uswtuxm5ba6zahdwh5atxhcsy

Table of Contents

2020 IEEE Transactions on Signal Processing  
Hsue 1776 Learning Nonnegative Factors From Tensor Data: Probabilistic Modeling and Inference Algorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  ...  So 5244 Linear Multiple Low-Rank Kernel Based Stationary GPs Regression for Time Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  ... 
doi:10.1109/tsp.2020.3042287 fatcat:nh7viihaozhd7li3txtadnx5ui

Tensor Completion via Convolutional Sparse Coding Regularization [article]

Zhebin Wu, Tianchi Liao, Chuan Chen, Cong Liu, Zibin Zheng, Xiongjun Zhang
2021 arXiv   pre-print
Tensor data often suffer from missing value problem due to the complex high-dimensional structure while acquiring them.  ...  To complete the missing information, lots of Low-Rank Tensor Completion (LRTC) methods have been proposed, most of which depend on the low-rank property of tensor data.  ...  And this is different from the Deep learning (CNN etc.) based approaches, where the statistical feature should be learned from lots of samples.  ... 
arXiv:2012.00944v2 fatcat:2exqpdoiw5hutekf63ppdwvi4e

MultiBench: Multiscale Benchmarks for Multimodal Representation Learning [article]

Paul Pu Liang, Yiwei Lyu, Xiang Fan, Zetian Wu, Yun Cheng, Jason Wu, Leslie Chen, Peter Wu, Michelle A. Lee, Yuke Zhu, Ruslan Salakhutdinov, Louis-Philippe Morency
2021 arXiv   pre-print
Learning multimodal representations involves integrating information from multiple heterogeneous sources of data.  ...  MultiBench provides an automated end-to-end machine learning pipeline that simplifies and standardizes data loading, experimental setup, and model evaluation.  ...  This is especially true for imperfections in the image modality. We believe that multimodal models are able to successfully rely on the other modality when one is  ... 
arXiv:2107.07502v2 fatcat:ls47dr7lpfhkbfry4r6dtqjtua

Tomographic image reconstruction using training images

Sara Soltani, Martin S. Andersen, Per Christian Hansen
2017 Journal of Computational and Applied Mathematics  
ii Summary (English) Reducing X-ray exposure while maintaining the image quality is a major challenge in computed tomography (CT); since the imperfect data produced from the few view and/or low intensity  ...  both with matrix and tensor representations of the training images.  ...  The regularization term C sum + C * results in coefficient tensors that are simultaneously low rank and sparse.  ... 
doi:10.1016/j.cam.2016.09.019 fatcat:o5b6je3strcybirrgxkifdk3w4

A Flexible Lossy Depth Video Coding Scheme Based on Low-rank Tensor Modelling and HEVC Intra Prediction for Free Viewpoint Video [article]

Mansi Sharma, Santosh Kumar
2021 arXiv   pre-print
Tensor factorization into a set of factor matrices following CANDECOMP PARAFAC (CP) decomposition via alternating least squares give a low-rank approximation of the scene geometry.  ...  The results demonstrate proposed approach achieves significant rate gains by efficiently compressing depth planes in low-rank approximated representation.  ...  The idea of lossy compression is inspired from low-rank modelling by tensor approximation that represents the high-dimensional depth data more compactly.  ... 
arXiv:2104.04678v1 fatcat:qjadbghnqzckjcjkxilqp5tbve

Time Series Forecasting via Learning Convolutionally Low-Rank Models [article]

Guangcan Liu
2021 arXiv   pre-print
Extensive experiments on 100,452 real-world time series from Time Series Data Library (TSDL) and M4 Competition (M4) demonstrate the superior performance of LbCNNM.  ...  Equipped with this learning method and some elaborate data argumentation skills, LbCNNM not only can handle well the major components of time series (including trends, seasonality and dynamics), but also  ...  We shall try to break through these limits via learning a proper representation for the target y.  ... 
arXiv:2104.11510v4 fatcat:utdr7llhhbgfnmd4kzq4jkqzxm

Learning Hawkes Processes from Short Doubly-Censored Event Sequences [article]

Hongteng Xu, Dixin Luo, Hongyuan Zha
2017 arXiv   pre-print
Experiments on both synthetic and real-world data demonstrate that the proposed data synthesis method improves learning results indeed for both time-invariant and time-varying Hawkes processes.  ...  Given SDC event sequences observed in a variety of time intervals, we propose a sampling-stitching data synthesis method --- sampling predecessors and successors for each SDC event sequence from potential  ...  This work is supported in part via NSF DMS-1317424, NIH R01 GM108341, NSFC 61628203, U1609220 and the Key Program of Shanghai Science and Technology Commission 15JC1401700.  ... 
arXiv:1702.07013v2 fatcat:zalknfrnfndqnhmfh4a2mnifum

2019 Index IEEE Transactions on Signal Processing Vol. 67

2019 IEEE Transactions on Signal Processing  
., +, TSP Feb. 15, 2019 1075-1087 Nonconvex Robust Low-Rank Tensor Reconstruction via an Empirical Bayes Method.  ...  ., +, TSP June 15, 2019 3287-3299 Nonconvex Robust Low-Rank Tensor Reconstruction via an Empirical Bayes Method.  ... 
doi:10.1109/tsp.2020.2968163 fatcat:dvvpqntb2rc2bjed5nnk4xora4

Deep Learning Methods for Solving Linear Inverse Problems: Research Directions and Paradigms [article]

Yanna Bai, Wei Chen, Jie Chen, Weisi Guo
2020 arXiv   pre-print
In this paper, we present a comprehensive survey of the recent progress in the development of deep learning for solving various linear inverse problems.  ...  We review how deep learning methods are used in solving different linear inverse problems, and explore the structured neural network architectures that incorporate knowledge used in traditional methods  ...  Dictionary learning denotes a LIP whose linear operator A and its representation m are learned from the observed data d, which exists in many applications such as image classification [47] , outliers  ... 
arXiv:2007.13290v2 fatcat:kqoerts77nftbl32fctx3za2me

Perspectives on Machine Learning-augmented Reynolds-averaged and Large Eddy Simulation Models of Turbulence [article]

Karthik Duraisamy
2021 arXiv   pre-print
Techniques to promote model-consistent training, and to avoid the requirement of full fields of direct numerical simulation data are detailed.  ...  Thus, machine learning should be viewed as one tool in the turbulence modeler's toolkit.  ...  Acknowledgements The author acknowledges support from the NSF CBET program (#1507928, Monitor: Dr. Ron  ... 
arXiv:2009.10675v3 fatcat:7n4z4nrfgbffrgomhpo5344sr4

Seq2Tens: An Efficient Representation of Sequences by Low-Rank Tensor Projections [article]

Csaba Toth, Patric Bonnier, Harald Oberhauser
2021 arXiv   pre-print
Sequential data such as time series, video, or text can be challenging to analyse as the ordered structure gives rise to complex dependencies.  ...  To address the innate computational complexity of high degree tensors, we use compositions of low-rank tensor projections.  ...  ACKNOWLEDGEMENTS CT is supported by the "Mathematical Institute Award" from the Mathematical Institute at the University of Oxford.  ... 
arXiv:2006.07027v2 fatcat:sbthoxojlzcn5anniqt5spfop4

Machine learning models of plastic flow based on representation theory [article]

Reese E. Jones, Jeremy A. Templeton, Clay M. Sanders, Jakob T. Ostien
2018 arXiv   pre-print
We use machine learning (ML) to infer stress and plastic flow rules using data from repre- sentative polycrystalline simulations.  ...  With these developments, we enable rapid model building in real-time with experiments, and guide data collection and feature discovery.  ...  Note loading in (d) is at a different rate from training data. Table 1 : 1 Time integration algorithm with adaptive time-stepping.  ... 
arXiv:1809.00267v1 fatcat:7cz3mvn66rendj3oi24w2gy7ji
« Previous Showing results 1 — 15 out of 822 results