Filters








6,266 Hits in 7.0 sec

Linearly constrained reconstruction of functions by kernels with applications to machine learning

R. Schaback, J. Werner
2006 Advances in Computational Mathematics  
By standard arguments of optimization theory, the solutions will take a simple form, based on the data related to the active constraints, called support vectors in the context of machine learning, The  ...  This paper investigates the approximation of multivariate functions from data via linear combinations of translates of a positive definite kernel from a reproducing kernel Hilbert space.  ...  Thus it is a reasonable idea to link point sets like Y ± to function spaces S for s, and this can be done by the "kernel trick" of learning theory that maps points y to functions K(·, y) in "feature space  ... 
doi:10.1007/s10444-004-7616-1 fatcat:ihwjbnpdijdotlzid4zukby6zi

Deep Learning Methods for Parallel Magnetic Resonance Image Reconstruction [article]

Florian Knoll, Kerstin Hammernik, Chi Zhang, Steen Moeller, Thomas Pock, Daniel K. Sodickson, Mehmet Akcakaya
2019 arXiv   pre-print
Following the success of deep learning in a wide range of applications, neural network-based machine learning techniques have received interest as a means of accelerating magnetic resonance imaging (MRI  ...  A number of ideas inspired by deep learning techniques from computer vision and image processing have been successfully applied to non-linear image reconstruction in the spirit of compressed sensing for  ...  Machine learning reconstruction approaches also come with a number of drawbacks when compared to classic constrained parallel imaging.  ... 
arXiv:1904.01112v1 fatcat:a5lxrnlwzzb5zjhut6znhpaxy4

Fast Fault Diagnosis in Industrial Embedded Systems Based on Compressed Sensing and Deep Kernel Extreme Learning Machines

Nanliang Shan, Xinghua Xu, Xianqiang Bao, Shaohua Qiu
2022 Sensors  
Inspired by the ideas of compressed sensing (CS) and deep extreme learning machines (DELM), a data-driven general method is proposed for fast fault diagnosis.  ...  With the complexity and refinement of industrial systems, fast fault diagnosis is crucial to ensuring the stable operation of industrial equipment.  ...  Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/s22113997 fatcat:d7jpcgzu3vaf5gefog7y3im7wq

Transductive inference & kernel design for object class segmentation

Dinh-Phong Vo, Hichem Sahbi
2012 2012 19th IEEE International Conference on Image Processing  
Our approach is based on the minimization of an energy function mixing i) a reconstruction term that factorizes a matrix of input data as a product of a learned dictionary and a learned kernel map ii)  ...  Transductive inference techniques are nowadays becoming standard in machine learning due to their relative success in solving many real-world applications.  ...  In that context, transductive versions of SVMs were also introduced [13, 11] ; they build decision functions by optimizing the parameters of a learning model together with the labels of the unlabeled  ... 
doi:10.1109/icip.2012.6467324 dblp:conf/icip/VoS12 fatcat:fisj3lmhufhvhgh3aeo53w22h4

Deep Probabilistic NMF Using Denoising Autoencoders

Satwik Bhattamishra
2018 International Journal of Machine Learning and Computing  
However, the algorithm is sensitive to noise and assumes that the signals in the data can be linearly reconstructed.  ...  The proposed model reduces the data to a lower dimensional manifold to get a more meaningful representation and takes into account the noisy nature of the data to improve the clustering performance of  ...  To minimize the overall loss function (13) with matrices and fixed, we need the derivatives of the function with respect to .  ... 
doi:10.18178/ijmlc.2018.8.1.662 fatcat:wakeiyb2rrbrhiv23i4ufunqne

Locality-Constrained Multi-Task Joint Sparse Representation for Image Classification

Lihua GUO
2013 IEICE transactions on information and systems  
Experiments on several challenging object classification datasets show that our proposed algorithm is better than the MTJSR, and is competitive with the state-of-the-art multiple kernel learning methods  ...  Lihua GUO †a) , Member SUMMARY In the image classification applications, the test sample with multiple man-handcrafted descriptions can be sparsely represented by a few training subjects.  ...  Acknowledgments This work is supported in part by the Guangzhou Science and Technology Plan Project(2012J2200010), the Fundamental Research Funds for the Central Universities of SCUT, the National Science  ... 
doi:10.1587/transinf.e96.d.2177 fatcat:jk6hz47yuzartntq4s37xze5lm

Stealing Neural Networks via Timing Side Channels [article]

Vasisht Duddu, Debasis Samanta, D Vijay Rao, Valentina E. Balas
2019 arXiv   pre-print
It is observed that it is possible to reconstruct substitute models with test accuracy close to the target models and the proposed approach is scalable and independent of type of Neural Network architectures  ...  In this paper, a black box Neural Network extraction attack is proposed by exploiting the timing side channels to infer the depth of the network.  ...  ACKNOWLEDGMENT The authors would like to thank Virat Shejwalkar(University of Massachusetts Amherst), Sasikumar Murakonda(National University of Singapore) and the reviewers for their feedback and helpful  ... 
arXiv:1812.11720v4 fatcat:hts4m64pabh37fp2jalgkoysu4

An Efficient FPGA-Based Hardware Accelerator for Convex Optimization-Based SVM Classifier for Machine Learning on Embedded Platforms

Srikanth Ramadurgam, Darshika G. Perera
2021 Electronics  
We incorporate suitable mathematical kernels and decomposition methods to systematically solve the convex optimization for machine learning applications with a large volume of data.  ...  on different platforms, and can be utilized for various machine learning applications.  ...  Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/electronics10111323 fatcat:tkncvfbl5zfxjjz7lmf3tt6uwi

Probabilistic local reconstruction for k-NN regression and its application to virtual metrology in semiconductor manufacturing

Seung-kyung Lee, Pilsung Kang, Sungzoon Cho
2014 Neurocomputing  
The "locally linear reconstruction" (LLR) provides a principled and k-insensitive way to determine the weights of k-nearest neighbor (k-NN) learning.  ...  LLR, however, does not provide a confidence interval for the k neighbors-based reconstruction of a query point, which is required in many real application domains.  ...  His research interests include instance-based learning, kernel learning machines, novelty detection, graph-based learning, learning algorithms in class imbalance and their applications such as keystroke  ... 
doi:10.1016/j.neucom.2013.10.001 fatcat:2lunzii2xjgyxhcwhghbn24ig4

Application of Support Vector Machines to Accelerate the Solution Speed of Metaheuristic Algorithms

Shiyou Yang, Q.H. Liu, Junwei Lu, S.L. Ho, Guangzheng Ni, Peihong Ni, Suming Xiong
2009 IEEE transactions on magnetics  
The support vector machine (SVM) is proposed as a response surface model to accelerate the solution speed of metaheuristic algorithms in solving inverse problems.  ...  Primary numerical results are reported to demonstrate the feasibility, performance, and robustness of the proposed SVM based response surface model for solving both mathematical functions and engineering  ...  SVM REGRESSION AND ITS APPLICATION A. SVM Regression The SVM developed by Vapnik and his coworkers [1] is a novel learning machine for classification and regression problems.  ... 
doi:10.1109/tmag.2009.2012690 fatcat:l75snapbq5eazjvmb2kmt423om

An user preference information based kernel for SVM active learning in content-based image retrieval

Hua Xie, Antonio Ortega
2004 Proceedings of the 6th ACM SIGMM international workshop on Multimedia information retrieval - MIR '04  
However, not much attention has been to paid to the design of novel kernel functions specifically tailored for relevance feedback problems and traditional kernels have been directly used in these applications  ...  kernels for SVMbased active learning in relevance feedback applications.  ...  By introducing Lagrange multipliers, this then leads to maximizing the Lagrangian objective function (3) with respect to positive Lagrange multipliers αi, i = 1, · · · , L, subject to constrains P i αiyi  ... 
doi:10.1145/1026711.1026713 dblp:conf/mir/XieO04 fatcat:jwdbuwvunra4rbkkqhjzsoe25q

Support Vector Machines for uncertainty region detection [chapter]

Gian Paolo Drago, Marco Muselli
2002 Perspectives in Neural Computing  
As in the Support Vector Machine approach, kernel functions can be used to generalize the proposed technique, so as to detect uncertainty regions with nonlinear boundaries.  ...  To this aim a modified version of the algorithm for the Generalized Optimal Hyperplane is shown to be effective.  ...  The application of ASVM, adopting the inner product as kernel leads to the (white) uncertainty regions U n depicted in Fig. 1b .  ... 
doi:10.1007/978-1-4471-0219-9_8 dblp:conf/wirn/DragoM01a fatcat:wdlsa7jbdzba7mw62gn4ihgjfa

Object correspondence as a machine learning problem

Bernhard Schölkopf, Florian Steinke, Volker Blanz
2005 Proceedings of the 22nd international conference on Machine learning - ICML '05  
We propose machine learning methods for the estimation of deformation fields that transform two given objects into each other, thereby establishing a dense point to point correspondence.  ...  Our system, which contains little engineering or domain knowledge, delivers state of the art performance. We present application results including close to photorealistic morphs of 3D head models.  ...  A large and important area where machine learning applications are relatively sparse, however, is the field of computer graphics.  ... 
doi:10.1145/1102351.1102449 dblp:conf/icml/ScholkopfSB05 fatcat:7j57d5smr5djnp5nmipxgaccpe

Comparison of sparse coding and kernel methods for histopathological classification of gliobastoma multiforme

Ju Han, Hang Chang, Leandro Loss, Kai Zhang, Fredrick L. Baehner, Joe W. Gray, Paul Spellman, Bahram Parvin
2011 2011 IEEE International Symposium on Biomedical Imaging: From Nano to Macro  
Experiments are designed to learn dictionaries, through sparse coding, and to train classifiers through kernel methods with normal, necorotic, apoptotic, and tumor with with characteristics of high cellularity  ...  Two different kernel methods of support vector machine (SVM) and kernel discriminant analysis (KDA) are used for comparative analysis.  ...  Acknowledgments Thanks to NIH for funding.  ... 
doi:10.1109/isbi.2011.5872505 pmid:23243485 pmcid:PMC3521607 dblp:conf/isbi/HanCLZBGSP11 fatcat:o6iitzswsfanjjzqvqz34z5h3y

Support Vectors Learning for Vector Field Reconstruction

Marcos Lage, Rener Castro, Fabiano Petronetto, Alex Bordignon, Geovan Tavares, Thomas Lewiner, Hélio Lopes
2009 2009 XXII Brazilian Symposium on Computer Graphics and Image Processing  
This paper proposes to formulate the unstructured vector field reconstruction and approximation through Machine-Learning.  ...  Reconstruction of a real 3D velocity field captured by a PIV device with sparse, irregular sampling: magnitude (left) and phase (right).  ...  ACKNOWLEDGEMENTS The authors thank CNPq, CAPES and FAPERJ for their support during the preparation of this work. We would like also to thank Prof. L.-F. Alzuguir for the PIV data-set.  ... 
doi:10.1109/sibgrapi.2009.20 dblp:conf/sibgrapi/LageCPBTLL09 fatcat:o5ptzmoom5d2phmny5idr65564
« Previous Showing results 1 — 15 out of 6,266 results