76,765 Hits in 4.0 sec

Randomized Algorithms for Large scale SVMs [article]

Vinay Jethava, Krishnan Suresh, Chiranjib Bhattacharyya, Ramesh Hariharan
2009 arXiv   pre-print
We propose a randomized algorithm for training Support vector machines(SVMs) on large datasets.  ...  Experiments done on synthetic and real life data sets demonstrate that the algorithm scales up existing SVM learners, without loss of accuracy.  ...  Classification This section uses results from random projections, and randomized algorithms for linear programming to develop a new algorithm for solving large scale SVM classification problems.  ... 
arXiv:0909.3609v1 fatcat:2zawse7p6bd6thjdg5agutoc6m

Hash-SVM: Scalable Kernel Machines for Large-Scale Visual Classification

Yadong Mu, Gang Hua, Wei Fan, Shih-Fu Chang
2014 2014 IEEE Conference on Computer Vision and Pattern Recognition  
This paper presents a novel algorithm which uses compact hash bits to greatly improve the efficiency of non-linear kernel SVM in very large scale visual classification problems.  ...  As a critical component of Hash-SVM, we propose a novel hashing scheme for arbitrary non-linear kernels via random subspace projection in reproducing kernel Hilbert space.  ...  Though parallel systems prove useful for scaling up large-scale learning techniques after properly tailoring the machine learning algorithms [3] , we target single machines with limited memory capacity  ... 
doi:10.1109/cvpr.2014.130 dblp:conf/cvpr/MuHFC14 fatcat:zgtbnhnkora6ppy7u2wav5v4fu

RANSAC-SVM for large-scale datasets

Kenji Nishida, Takio Kurita
2008 Pattern Recognition (ICPR), Proceedings of the International Conference on  
Support Vector Machines (SVMs), though accurate, are still difficult to solve large-scale applications, due to the computational and storage requirement.  ...  To relieve this problem, we propose RANSAC-SVM method, which trains a number of small SVMs for randomly selected subsets of training set, while tuning their parameters to fit SVMs to whole training set  ...  in large-scale problems.  ... 
doi:10.1109/icpr.2008.4761280 dblp:conf/icpr/WatanabeK08a fatcat:3vsi4gfdsfgurg3oopfxm2ee4y

Quantum-Enhanced Machine Learning for Covid-19 and Anderson Insulator Predictions [article]

Paul-Aymeric McRae, Michael Hilke
2020 arXiv   pre-print
Quantum Machine Learning (QML) algorithms to solve classifications problems have been made available thanks to recent advancements in quantum computation.  ...  For the computation, we used the 16 qubit IBM quantum computer. We find that the "quantum enhancement" is not generic and fails for more complex machine learning tasks.  ...  Quantum Support Vector Machines The quantum version of the SVM is best described as "quantum-assisted" or "quantum-enhanced" [19] in the sense that the algorithm is largely classical with certain operations  ... 
arXiv:2012.03472v1 fatcat:g3isjby7gnh6fpxe5jkivsqx5q

Dual Random Projection for Linear Support Vector Machine

Xi XI, Feng-qin ZHANG, Xiao-qing LI
2017 DEStech Transactions on Computer Science and Engineering  
However, the large-scale SVM has the problem of reduced classification accuracy after dimension reduction by random projection feature.  ...  Therefore, we propose a linear kernel support vector machine based on dual-random projection (drp-LSVM) for large-scale classification problem, which combines the duality recovery theory.  ...  In recent years, stochastic approximation algorithms have been widely used in large-scale machine learning, and the Random Projections (RP) method can quickly and effectively solve the dimensionality problem  ... 
doi:10.12783/dtcse/smce2017/12422 fatcat:ff4hhyqfcfbp7c2y4usk7gq2ee

Parallel multiclass stochastic gradient descent algorithms for classifying million images with very-high-dimensional signatures into thousands classes

Thanh-Nghi Do
2014 Vietnam Journal of Computer Science  
We extend the stochastic gradient descent (SGD) for support vector machines (SVM-SGD) in several ways to develop the new multiclass SVM-SGD for efficiently classifying large image datasets into many classes  ...  We propose (1) a balanced training algorithm for learning binary SVM-SGD classifiers, and (2) a parallel training process of classifiers with several multi-core computers/grid.  ...  [8] and [9] propose stochastic gradient descent algorithms for SVM (denoted by SVM-SGD) that shows the promising results for large-scale binary classification problems.  ... 
doi:10.1007/s40595-013-0013-2 fatcat:5cqdbgvkgjbmvd7ndvnifkiyxi

Preselection of support vector candidates by relative neighborhood graph for large-scale character recognition

Masanori Goto, Ryosuke Ishida, Seiichi Uchida
2015 2015 13th International Conference on Document Analysis and Recognition (ICDAR)  
We propose a pre-selection method for training support vector machines (SVM) with a large-scale dataset.  ...  Through large-scale handwritten digit pattern recognition experiments, we show that the proposed preselection method accelerates SVM training process 5-15 times faster without degrading recognition accuracy  ...  • We developed an efficient method for constructing RNGs. The computation time of the method scales with O(n 2 ) and the algorithm is memory-efficient.  ... 
doi:10.1109/icdar.2015.7333773 dblp:conf/icdar/GotoIU15 fatcat:d6wgrpnrjzb4hegqirexyju3zy


2017 International Journal of Applied Mathematics and Machine Learning  
For large sample classification model, it is necessary to explore new SVM training algorithm for large-scale data set.  ...  Based on practical examples, three kernel parameter optimization methods of Support Vector Machines (SVM), i.e., grid method, particle swarm optimization, and genetic algorithm, are compared.  ...  For the classification model of large samples, the SVM training algorithm for large scale data sets are needed.  ... 
doi:10.18642/ijamml_7100121908 fatcat:4nwoqm4bzrabfjurpygieq6qqy

Gender Recognition Using a Min-Max Modular Support Vector Machine with Equal Clustering [chapter]

Jun Luo, Bao-Liang Lu
2006 Lecture Notes in Computer Science  
Through task decomposition and module combination, minmax modular support vector machines (M 3 -SVMs) can be successfully used for different pattern classification tasks.  ...  Based on an equal clustering algorithm, M 3 -SVMs can divide the training data set of the original problem into several subsets with nearly equal number of samples, and combine them to a series of balanced  ...  Acknowledgements The authors thank Yi-Min Wen for providing the source programme of equal clustering algorithm.  ... 
doi:10.1007/11760023_31 fatcat:xt2ryvg3gzd2bmh2ixaql3uika

Comments on "A Parallel Mixture of SVMs for Very Large Scale Problems"

Xiaomei Liu, Lawrence O. Hall, Kevin W. Bowyer
2004 Neural Computation  
This approach has the advantage that the required computation scales well to very large data sets.  ...  Collobert et. al. recently introduced a novel approach to using a neural network to provide a class prediction from an ensemble of support vector machines (SVMs).  ...  It is conjectured that the time cost of a parallel mixture of SVMs is sub-quadratic with the number of training patterns for large scale problems.  ... 
doi:10.1162/089976604323057416 pmid:15165393 fatcat:t7if6syz45bilmgejxhcjztijm

Support Vector Machines Trained with Evolutionary Algorithms Employing Kernel Adatron for Large Scale Classification of Protein Structures

Nancy Arana-Daniel, Alberto A. Gallegos, Carlos López-Franco, Alma Y. Alanís, Jacob Morales, Adriana López-Franco
2016 Evolutionary Bioinformatics  
CiTATion: Daniel et al. support vector machines trained with Evolutionary algorithms Employing Kernel Adatron for Large Scale Classification of Protein Structures.  ...  Most state of the art approaches for large-scale learning use traditional optimization methods, such as quadratic programming or gradient descent, which makes the use of evolutionary algorithms for training  ...  Acknowledgment We are very thankful to the projects SEP CONACYT CB-258068 and CB-256769 for supporting this work. Author contributions Conceived and proposed the initial idea of the work: NAD.  ... 
doi:10.4137/ebo.s40912 pmid:27980384 pmcid:PMC5140013 fatcat:ipq7hftjq5eyxat4jmlga5r5tq

Multiclass relevance units machine: benchmark evaluation and application to small ncRNA discovery

Mark Menor, Kyungim Baek, Guylaine Poisson
2013 BMC Genomics  
Furthermore, McRUM is computationally more efficient than the SVM, which is an important factor for large-scale analysis. Conclusions: We have proposed McRUM, a multiclass extension of binary CRUM.  ...  McRUM with Naïve decoding algorithm is computationally efficient in run-time and its predictive performance is comparable to the well-known SVM, showing its potential in solving large-scale multiclass  ...  Declarations The publication costs for this article were funded by grant number  ... 
doi:10.1186/1471-2164-14-s2-s6 pmid:23445533 pmcid:PMC3582431 fatcat:c4yk33okmnhdbp7tckza3ru3sa

b-Bit Minwise Hashing for Large-Scale Linear SVM [article]

Ping Li and Joshua Moore and Christian Konig
2011 arXiv   pre-print
Our technique is particularly useful when the data can not fit in memory, which is an increasingly critical issue in large-scale machine learning.  ...  Interestingly, our proof for the positive definiteness of the b-bit minwise hashing kernel naturally suggests a simple strategy to integrate b-bit hashing with linear SVM.  ...  to apply it to many large-scale learning problems.  ... 
arXiv:1105.4385v1 fatcat:bshphzwwazamjgcs3rn2pnkgwm

Fast and scalable polynomial kernels via explicit feature maps

Ninh Pham, Rasmus Pagh
2013 Proceedings of the 19th ACM SIGKDD international conference on Knowledge discovery and data mining - KDD '13  
Approximation of non-linear kernels using random feature mapping has been successfully employed in large-scale data analysis applications, accelerating the training of kernel machines.  ...  Empirically, Tensor Sketching achieves higher accuracy and often runs orders of magnitude faster than the state-of-the-art approach for large-scale real-world datasets.  ...  on 4 large-scale datasets.  ... 
doi:10.1145/2487575.2487591 dblp:conf/kdd/PhamP13 fatcat:ixrmgagtsbfcfifa6ysrgjzmle

Solving Linear SVMs with Multiple 1D Projections

Johannes Schneider, Jasmina Bogojeska, Michail Vlachos
2014 Proceedings of the 23rd ACM International Conference on Conference on Information and Knowledge Management - CIKM '14  
We present a new methodology for solving linear Support Vector Machines (SVMs) that capitalizes on multiple 1D projections.  ...  Our solution adapts on methodologies from random projections, exponential search, and coordinate descent.  ...  This makes it an attractive method for solving large-scale linear SVM problems. EMPIRICAL EVALUATION Here we evaluate the runtime and accuracy of the proposed methods.  ... 
doi:10.1145/2661829.2661994 dblp:conf/cikm/SchneiderBV14 fatcat:s3or76ldzvhdzdwt4jwh3khmfy
« Previous Showing results 1 — 15 out of 76,765 results