Sparse Least-Squares Support Vector Machines via Accelerated Segmented Test: A dual approach

Oliveira, Saulo A. F.; Gomes, João P. P.; Neto, Ajalmar R. da Rocha

doi:10.1016/j.neucom.2018.08.081

Cited by 12 publications

(10 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, it turns out that FAST formulation only deals with image data, i.e., two dimensional samples that are uniformly spaced in a grid. To overcome such limitations, [4] extended FAST so that we can apply it to highdimensional inputs in a straightforward way. Although, the authors did not give it a proper name in [4], here we call it Class-Corner Instance Selection (CCIS).…”

Section: Contribution 1: Class-corner Instance Selectionmentioning

confidence: 99%

“…To overcome such limitations, [4] extended FAST so that we can apply it to highdimensional inputs in a straightforward way. Although, the authors did not give it a proper name in [4], here we call it Class-Corner Instance Selection (CCIS).…”

Section: Contribution 1: Class-corner Instance Selectionmentioning

confidence: 99%

“…where D = {(x i , y i )} N i=1 . We discussed in [4] that by adopting P = 0 and default values for K and P , we derived feasible subsets that share this class-corner feature. We validated our first contribution alongside the following one.…”

Section: Contribution 1: Class-corner Instance Selectionmentioning

confidence: 99%

“…This thesis unfolds four contributions. The first two were evaluated together in [4], while the third one is presented in [5]. The last part has not been published yet.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

On Model Complexity Reduction in Instance-Based Learners

Oliveira

Neto²,

Gomes

2022

Anais Estendidos Do XXXV Conference on Graphics, Patterns and Images (SIBGRAPI Estendido 2022)

View full text Add to dashboard Cite

Instance-based learners habitually adopt instance selection techniques to reduce complexity and avoid overfitting. Such learners’ most recent and well-known formulations seek to impose some sparsity in their training and prediction structure alongside regularization to meet such a result. Due to the variety of such instance-based learners, we will draw attention to the Least-Squares Support Vector Machines and Minimal Learning Machines because they embody additional information beyond the stored instances to perform predictions. Later, this thesis proposes variants constraining candidate solutions within a specific functional space where we avoid overfitting and reduce model complexity. The central core of such variants is related to penalizing samples with a specific condition during learning. For regressors, we adopted strategies based on random and observed linearity conditions related to the data. At the same time, we borrowed definitions from the computer vision field for classification tasks to derive a concept we call the classcorner relationship (in which we designed an instance selection algorithm). In the Least-Squares Support Vector Machines context, this thesis follows the pruning fashion by adopting the samples that share such a class-corner relationship. As for the Minimal Learning Machine model, this thesis introduces a new proposal called the Lightweight Minimal Learning Machine, a faster model for out-of-sample prediction due to the reduced number of computations inherent in the original proposal’s multilateration process. Another remarkable feature is that it derives a unique solution when other formulations rely on overdetermined systems.

show abstract

Section: Contribution 1: Class-corner Instance Selectionmentioning

confidence: 99%

Section: Contribution 1: Class-corner Instance Selectionmentioning

confidence: 99%

Section: Contribution 1: Class-corner Instance Selectionmentioning

confidence: 99%

“…This thesis unfolds four contributions. The first two were evaluated together in [4], while the third one is presented in [5]. The last part has not been published yet.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

On Model Complexity Reduction in Instance-Based Learners

Oliveira

Neto²,

Gomes

2022

Anais Estendidos Do XXXV Conference on Graphics, Patterns and Images (SIBGRAPI Estendido 2022)

View full text Add to dashboard Cite

show abstract

“…For comparison with the purely LSSVM model, the resultant values of 822.5850 and 6021.7017 for and 2 , respectively, were used. 39 A 6-7-1 structure with a learning rate of 0.003 and momentum factor of 0.9 using Bayesian regularization (BR) algorithm was selected for the BP-NN according to Yang et al's method. 40 The models of the dyeing knit cotton fabric using cold pad-batch process were established by training with 80 groups of dyeing experiments (Supplemental Table S3).…”

Section: Designing Of Experiments For Dyeing Model By Pso-lssvmmentioning

confidence: 99%

K/S value prediction of cotton fabric using PSO-LSSVM

et al. 2020

Textile Research Journal

View full text Add to dashboard Cite

Cotton is one of the world’s most common natural clothing materials. It is dyed mainly using the exhaustion, cold pad-batch, and pad-dry-pad-steam dyeing methods. The K/S value, an important index for measuring the depth of color, of cotton fabric dyed with reactive dyes is greatly influenced by various factors of the dyeing process. In this study, three models were developed incorporating least squares support vector machine (LSSVM) to predict the K/S values of dyed cotton fabrics, while particle swarm optimization (PSO) was applied to optimize and tune the parameters of the LSSVM model (PSO-LSSVM). Model inputs include dye concentration and process conditions, which are both easily obtainable variables. The K/S values from the PSO-LSSVM model are consistent with actual measured K/S values of dyed cotton fabrics. Moreover, a comparison among PSO-LSSVM, LSSVM and back propagation neural network results shows the superiority of the PSO-LSSVM approach. Results of this work indicate that a PSO-LSSVM model is a powerful tool for predicting the K/S value in cotton fabric dyed with reactive dye and thus a means to improve production processes and reduce costs.

show abstract

Image Registration in Medical Robotics and Intelligent Systems: Fundamentals and Applications

Liu

Singh

Al’Aref

et al. 2019

Advanced Intelligent Systems

View full text Add to dashboard Cite

Medical image registration, by transforming two or more sets of imaging data into one coordinate system, plays a central role in medical robotics and intelligent systems from diagnostics and surgical planning to real‐time guidance and postprocedural assessment. Recent advances in medical image registration have made a significant impact in orthopedic, neurological, cardiovascular, and oncological applications.The recent literature in medical image registration is reviewed, providing a discussion of their fundamentals and applications. Within each section, the registration techniques are introduced, classifying each method based on their working mechanisms, and discussing their benefits and limitations are discussed. Recently, machine learning has had an important impact on the field of image registration, yielding novel methods and unprecedented speed. The validation of registration methods, however, remains a challenge due to the lack of reliable ground truth. Medical image registration will continue to make significant impacts in the area of advanced medical imaging, as the fusion/combination of multimodal images and advanced visualization technology become more widespread.

show abstract

Sparse Least-Squares Support Vector Machines via Accelerated Segmented Test: A dual approach

Cited by 12 publications

References 19 publications

On Model Complexity Reduction in Instance-Based Learners

On Model Complexity Reduction in Instance-Based Learners

K/S value prediction of cotton fabric using PSO-LSSVM

Image Registration in Medical Robotics and Intelligent Systems: Fundamentals and Applications

Contact Info

Product

Resources

About