A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
Filters
Expectation-Maximization Tensor Factorization for Practical Location Privacy Attacks
2017
Proceedings on Privacy Enhancing Technologies
Preserved Fulltext
We apply our learning methods to a de-anonymization attack and a localization attack, and evaluate them using three real datasets. ...
missing locations (E-step) and computing personalized transition matrices via tensor factorization (M-step). ...
Acknowledgement: The author would like to thank Jacob Schuldt (AIST), Atsunori Kanemura (AIST), and Hideitsu Hino (University of Tsukuba) for technical comments on this paper. ...
doi:10.1515/popets-2017-0042
dblp:journals/popets/Murakami17
fatcat:6iv736nw3fdixa6votvrytljca
Desynchronization resilient video fingerprinting via randomized, low-rank tensor approximations
2011
2011 IEEE 13th International Workshop on Multimedia Signal Processing
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2013; you can also visit the original URL.
The file type is application/pdf.
In this paper, we model videos as order-3 tensors and use multilinear subspace projections, such as a reduced rank parallel factor analysis (PARAFAC) to construct video hashes. ...
The most significant gains are seen for the difficult attacks of spatial (e.g. geometric) as well as temporal (random frame dropping) desynchronization. ...
Our proposed algorithm implemented PARAFAC tensor factorization using tensor toolbox provided by Sandia National Labs [14] . ...
doi:10.1109/mmsp.2011.6093778
dblp:conf/mmsp/LiM11
fatcat:hneqoxh4afhhdmhjjuwuue5vkm
Robust Image Hashing Based on Multi-view Feature Representation and Tensor Decomposition
2022
Journal of Information Hiding and Multimedia Signal Processing
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Then, the popular algorithm Tucker Decomposition (TD) is applied to decompose the tensor into core tensor and factor matrices. Finally, a robust hash code is constructed by using factor matrices. ...
This paper proposes a Robust Image Hashing Based on Multi-view Feature Representation and Tensor Decomposition. ...
Finally, since the factor matrix can fully express the internal structure of the original tensor while ensuring robustness, we construct hash sequences with factor matrices and encoded them into a compact ...
dblp:journals/jihmsp/ShangDW22
fatcat:lwcdaxy5z5hf5nhrth3vjvyacy
PairFac
2016
Proceedings of the 25th ACM International on Conference on Information and Knowledge Management - CIKM '16
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
We also demonstrate the advantages of our approach through case studies with real-world traffic sensor data and social media streams surrounding the 2015 terrorist attacks in Paris. ...
PairFac utilizes discriminant tensor analysis to automatically discover the impact of a major event from rich human behavioral data. ...
Any opinions, findings, and conclusions or recommendations expressed in this material do not necessarily reflect the views of the funding sources. ...
doi:10.1145/2983323.2983837
dblp:conf/cikm/WenLP16
fatcat:jr362q6ux5fpdamkmrrzoxyc6a
Convolutional Neural Networks with Transformed Input based on Robust Tensor Network Decomposition
[article]
2018
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
of different adversarial attacks including global and localized attacks, and the efficacy of different adversarial defenses based on input transformation. ...
The theory is extended to analyze higher-order tensors using tensor-train SVD (TT-SVD); it helps to explain the level of susceptibility of different datasets to adversarial attacks, the structural similarity ...
Tensor-Train (TT) decomposition [55, 54] decomposes a given tensor into a matrix, followed by a series of threemode "transfer" core tensors, and finally ended by a matrix. ...
arXiv:1812.02622v2
fatcat:dkdjfsanlbbqlkbh5two34sebm
Multiple Strategies Differential Privacy on Sparse Tensor Factorization for Network Traffic Analysis in 5G
2022
IEEE Transactions on Industrial Informatics
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Meanwhile, Sparse Tensor Factorization (STF) is a useful tool for dimension reduction to analyze High-Order, High-Dimension, and Sparse Tensor (HOHDST) data which is transmitted on 5G Internetof-things ...
MDPSTF comprises three Differential Privacy (DP) mechanisms, i.e., ε− DP, Concentrated DP (CDP), and Local DP (LDP). Furthermore, the theoretical proof of privacy bound is presented. ...
Output: x Factor matrices A ∈ R I×R , B ∈ R J×R , and C ∈ R K×R , y Noise factor matrix  ∈ R I×R and recovered tensor X , z Noise factor matrix Ā ∈ R I×R and recovered tensor X , 1 xInitialize A ∈ R I ...
doi:10.1109/tii.2021.3082576
fatcat:koj3fdluizdlla6e4f2iujw5ky
TensorShield: Tensor-based Defense Against Adversarial Attacks on Images
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Our tensor-based defense mechanism outperforms the SLQ method from Shield by 14% against FastGradient Descent (FGSM) adversarial attacks, while maintaining comparable speed. ...
Subtle and imperceptible perturbations of the data are able to change the result of deep neural networks. ...
Tensor-based Defense Mechanism In this section, we briefly describe concepts and notations used in the paper. A tensor, denoted by X, is a multidimensional matrix. ...
arXiv:2002.10252v1
fatcat:dpwlxzmoafbftexe3fxj652poa
Tensor Dropout for Robust Learning
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
While standard CNNs use matrix computations, we study tensor layers that involve higher-order computations and provide better inductive bias. ...
However, they have limited generalization ability to data outside the training domain, and a lack of robustness to noise and adversarial attacks. ...
This can then be used to sketch the factors U (k) of the decomposition asŨ (k) = U (k) (M (k) ) be a sketch of factor matrix , and the core tensor G as G = G × 0 M (0) × · · · × N M (N ) . ...
arXiv:1902.10758v4
fatcat:hq2i5u56lncmzni5wvupajv7ke
Mining billion-scale tensors: algorithms and discoveries
2016
The VLDB journal
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
For analyzing a tensor, tensor decompositions are widely used in many data mining applications: detecting malicious attackers in network traffic logs with (source IP, destination IP, portnumber, timestamp ...
However, current tensor decomposition methods do not scale to large and sparse real-world tensors with millions of rows and columns and 'fibers'. ...
PARAFAC is useful for decomposing a tensor into rank-one tensors which form the latent factors; Tucker is more suitable for compressing tensors and examining relations between the latent factors. ...
doi:10.1007/s00778-016-0427-4
fatcat:wiyizdzhd5hbxdflnrsaw6foqi
MalSpot: Multi2 Malicious Network Behavior Patterns Analysis
[chapter]
2014
Lecture Notes in Computer Science
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2014; you can also visit the original URL.
The file type is application/pdf.
In this work, we develop MalSpot, multi-resolution and multi-linear (Multi 2 ) network analysis system in order to discover such malicious patterns, so that we can use them later for attack detection, ...
We designed and deployed MalSpot, which employs multi-linear analysis with different time resolutions, running on top of MapReduce (Hadoop), and we identify patterns across attackers, attacked institutions ...
IIS-1247489 and CNS-1314632 Research was sponsored by the Defense Threat Reduction Agency and was accomplished under contract No. HDTRA1-10-1-0120. ...
doi:10.1007/978-3-319-06608-0_1
fatcat:wav7bl7i2vfi5lhefvgamjww7q
Communication Efficient Generalized Tensor Factorization for Decentralized Healthcare Networks
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Other Versions
However, existing federated tensor factorization algorithms encounter the single-point-failure issue with the involvement of the central server, which is not only easily exposed to external attacks but ...
strategy designed for a generalized tensor factorization, which has the flexibility of modeling different tensor distribution with multiple kinds of loss functions. ...
, CTSA Award UL1TR002378, and Cisco Research University Award #2738379. ...
arXiv:2109.01718v2
fatcat:k6qivuxsizgebfmxat7v465dr4
High Dynamic Range Image Watermarking Based on Tucker Decomposition
2019
IEEE Access
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
strategy is used to balance the imperceptibility and robustness of the watermark. ...
Then, a Auto-Regressive prediction method is used to establish a local correlation model of the first feature map, so that watermark can be embedded in the first feature map according to the prediction ...
However, tensor decomposition can effectively overcome these problems. Currently, there are two commonly used factored forms of tensor decomposition. ...
doi:10.1109/access.2019.2935627
fatcat:pmtjy2g3erdfnetnznyjcwaqeu
Comprehensive Feature-based Robust Video Fingerprinting Using Tensor Model
[article]
2016
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Therefore, in the present study, we mine the assistance and consensus among different features based on tensor model, and present a new comprehensive feature to fully use them in the proposed video fingerprinting ...
A matching strategy used for narrowing the search is also proposed based on the core tensor. ...
A (n) is the factor matrix. Tensor has been applied to many fields, such as computer vision and signal processing. ...
arXiv:1601.07270v1
fatcat:3xu4pojk3nemhezmxzzvtnyjge
Error-Robust Distributed Denial of Service Attack Detection Based on an Average Common Feature Extraction Technique
2020
Sensors
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Furthermore, traditional machine learning-based intrusion detection systems (IDSs) often fail to efficiently detect such attacks when corrupted datasets are used for IDS training. ...
In addition, for error-free conditions, it is found that the proposed approach outperforms other related works, showing accuracy, detection rate and false alarm rate of 99.87%, 99.86% and 0.16%, respectively ...
C C C Weight tensor X :,n n-th dataset feature X X X [f] Filtered dataset tensor A r r-th factor matrix [X X X] (r) r-th mode unfolding matrix of X X X
Number of features along the r-th dimension X ...
doi:10.3390/s20205845
pmid:33081079
fatcat:rimt7oqnyfeklabqcx74lwfmnq
Detecting and Locating Passive Video Forgery Based on Low Computational Complexity Third-Order Tensor Representation
2021
Journal of Imaging
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Forgery affects video integrity and authenticity and has serious implications. For example, digital videos for security and surveillance purposes are used as evidence in courts. ...
Experimental results and comparisons show the superiority of the proposed scheme with a precision value of up to 99% in detecting and locating both types of attacks for static as well as dynamic videos ...
Acknowledgments: The authors thank Department of Information Technology, Faculty of Computers and Information Science, Mansoura University. ...
doi:10.3390/jimaging7030047
pmid:34460703
pmcid:PMC8321313
fatcat:djbnutbzvngmpfn63roaeoj2zq
« Previous
Showing results 1 — 15 out of 12,704 results