LIBSVM: A library for support vector machines

By Chih-Jen Lin. 2011

44,984 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Evaluation & Benchmarks

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Evaluation & Benchmarks

Citations in the corpus

Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition (2014, methodology): “But in practice the GPU implementations (such as cuda-convnet [3] and Caffe [35]) are preferably run on fixed input images.”
Remote Sensing Image Scene Classification: Benchmark and State of the Art (2017, methodology): “We implemented it with the LibSVM toolbox [176] and used the default setting in linear SVM (C=1) to tune”
Scikit-learn: Machine Learning in Python (2011, methodology): “While the package is mostly written in Python, it incorporates the C++ libraries LibSVM (Chang and Lin, 2001) and LibLinear (Fan et al., 2008) that provide reference implementations of SVMs and generalized linear models with compatible licenses.”
Learning Convolutional Neural Networks for Graphs (2016, methodology): “We performed 10 -fold cross-validation with LIB-SVM (Chang & Lin, 2011), using 9 folds for training and 1 for testing, and repeated the experiments 10 times.”
Discriminative Embeddings of Latent Variable Models for Structured Data (2016, neutral): “…for structured kernel methods, we tune the degree in {1, 2, . . . , 10} (for mismatch kernel, we also tune the maximum mismatch length in {1, 2, 3}) and train SVM classifier (Chang & Lin, 2001) on top, where the trade-off parameter C is also chosen in {0.01, 0.1, 1, 10} by cross validation.”
Parsing Algebraic Word Problems into Equations (2015, methodology): “For the local and global models, we use the LIBSVM package to train SVM classifiers (Chang and Lin, 2011) with RBF kernels that return likelihood estimates as the score.”
Dynamics of SGD with Stochastic Polyak Stepsizes: Truly Adaptive Variants and Convergence to Exact Solution (2022, methodology): “We study performance on three datasets: (1) a Synthetic Dataset, (2) The A1A dataset [6] and (3) the Breast Cancer dataset [10].”
Stochastic Polyak Step-size for SGD: An Adaptive Learning Rate for Fast Convergence (2020, methodology): “We experiment with four standard datasets: mushrooms, rcv1, ijcnn, and w8a from LIBSVM (Chang and Lin, 2011).”
Fast and Furious Convergence: Stochastic Second Order Methods under Interpolation (2019, methodology): “Real datasets: We also consider real datasets mushrooms, rcv1, and ijcnn from the LIBSVM repository (Chang and Lin, 2011), and use an 80 : 20 split for the training and test set, respectively.”
Painless Stochastic Gradient: Interpolation, Line-Search, and Convergence Rates (2019, methodology): “We experiment with four standard datasets: mushrooms, rcv1, ijcnn, and w8a from LIBSVM [15].”
CNN Features Off-the-Shelf: An Astounding Baseline for Recognition (2014, methodology): “2 +C X i max(1 y iwT x i;0) (1) When training an SVM for one class we used all the images containing an instance of that class as the positive samples and the rest as negative samples. We used libsvm [9] with the trade-off parameter set to C=5 for all classes and chosen by cross-validation on the tr”
Marginalized Denoising Autoencoders for Domain Adaptation (2012, methodology): “Glorot et al. (2011) demonstrate that using SDA-learned features in conjunction with linear SVM classiﬁers yields record performance on the benchmark tasks of sentiment analysis across different product domains (Blitzer et al., 2006).”
Distributed Fixed Point Methods with Compressed Iterates (2019, positive): “In this section we present numerical results that demonstrate the conclusions of the theoretical convergence results for Algorithms FPMCI (Algorithm 3) and VR-FPMCI (Algorithm 4) on LibSVM datasets [9].”
First Analysis of Local GD on Heterogeneous Data (2019, methodology): “3 Experiments To verify the theory, we run our experiments on logistic regression with `2 regularization and datasets taken from the LIBSVM library [2].”
Stealing Hyperparameters in Machine Learning (2018, background): “However, our attacks are also applicable to multi-class classiﬁcation algorithms such as multi-class support vector machine [11] and multi-class logistic regression [11] that use hyperparameters in their objective functions.”
Evasion and Hardening of Tree Ensemble Classifiers (2015, methodology): “L2) and RBF-SVM trained using LibLinear [16] and LibSVM [17] respectively.”
Analysis of classifiers’ robustness to adversarial perturbations (2015, methodology): “we use the efficient Liblinear (Fan et al, 2008) and LibSVM (Chang and Lin, 2011) implementations, and we fix the regularization parameters using a cross-validation procedure.”