High-Performance Parallel Support Vector Machine Training

Woodsend, Kristian; Gondzio, Jacek

doi:10.1007/978-0-387-09707-7_7

Cited by 28 publications

(29 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Then at the end of the multiplication, a single gather operation is required on the (m + 1) × (m + 1) matrix at each processor to form the matrix M and then factorize it. Implementation details are addressed in [33]. This could point the way forward for tackling large and complex data sets.…”

Section: Discussionmentioning

confidence: 99%

Exploiting separability in large-scale linear support vector machine training

Woodsend

Gondzio

2009

Comput Optim Appl

Self Cite

View full text Add to dashboard Cite

Linear support vector machine training can be represented as a large quadratic program. We present an efficient and numerically stable algorithm for this problem using interior point methods, which requires only O(n) operations per iteration. Through exploiting the separability of the Hessian, we provide a unified approach, from an optimization perspective, to 1-norm classification, 2-norm classification, universum classification, ordinal regression and -insensitive regression. Our approach has the added advantage of obtaining the hyperplane weights and bias directly from the solver. Numerical experiments indicate that, in contrast to existing methods, the algorithm is largely unaffected by noisy data, and they show training times for our implementation are consistent and highly competitive. We discuss the effect of using multiple correctors, and monitoring the angle of the normal to the hyperplane to determine termination.

show abstract

Section: Discussionmentioning

confidence: 99%

Exploiting separability in large-scale linear support vector machine training

Woodsend

Gondzio

2009

Comput Optim Appl

Self Cite

View full text Add to dashboard Cite

show abstract

“…First of all, the number of training examples was large enough to create a problem for our computational resources. The scaling of SVM to large data sets is indeed an active research area [2,7,18,19]. We turned our attention to a simple approach proposed by V.Vapnik et al in [11], called Cascade SVM.…”

Section: Tablementioning

confidence: 99%

Conformal prediction of biological activity of chemical compounds

Toccaceli

Nouretdinov

Gammerman

2017

Ann Math Artif Intell

View full text Add to dashboard Cite

The paper presents an application of Conformal Predictors to a chemoinformatics problem of predicting the biological activities of chemical compounds. The paper addresses some specific challenges in this domain: a large number of compounds (training examples), high-dimensionality of feature space, sparseness and a strong class imbalance. A variant of conformal predictors called Inductive Mondrian Conformal Predictor is applied to deal with these challenges. Results are presented for several non-conformity measures extracted from underlying algorithms and different kernels. A number of performance measures are used in order to demonstrate the flexibility of Inductive Mondrian Conformal Predictors in dealing with such a complex set of data. This approach allowed us to identify the most likely active compounds for a given biological target and present them in a ranking order.

show abstract

“…SVMs are powerful machine learning techniques for classification and regression [12]. In proposed method SVM is used for prediction of the electric field and magnetic field for increase in tower height.…”

Section: Prediction Of Electric and Magnetic Field For Increase In Tomentioning

confidence: 99%

Prediction of Electromagnetic Fields around High Voltage Transmission Lines

Kulkarni

Gandhare²

2017

Acta Tech Jaur

View full text Add to dashboard Cite

The electric and magnetic fields are present around the High Voltage Transmisssion Lines (HVTL) and reported to affect health of the workers working on these hotlines. The key parameter reportedly responsible for detrimental health effects, are internal induced fields in body. Induction of internal fields in different organs is heavily dependent on the external fields. Prediction of hazardous levels of external fields before measurement or in situations difficult for direct measurement will lead to identify the restrictive situations and working conditions for hotline workers. This work propose a method to model electric and magnetic field for different climbing routes using hybrid technique, formed by combining support vector machine (SVM) and neural network (NN) and also electric field and magnetic field values are predicted using NN for increase in tower height. The result shows the performance of proposed method for prediction of electric field and magnetic field for increase in tower height.

show abstract

High-Performance Parallel Support Vector Machine Training

Cited by 28 publications

References 8 publications

Exploiting separability in large-scale linear support vector machine training

Exploiting separability in large-scale linear support vector machine training

Conformal prediction of biological activity of chemical compounds

Prediction of Electromagnetic Fields around High Voltage Transmission Lines

Contact Info

Product

Resources

About