New algorithms parallelize all-in-one SVM training, improving text classification accuracy.
problem Training all-in-one SVMs is computationally infeasible for many classes.
method Developed distributed algorithms for two all-in-one SVM formulations.
result Superior accuracy on text classification data compared to one-vs.-rest SVMs.
Proposes SOVR loss to improve adversarial robustness by increasing logit margins.
problem Adversarial training's difficulty in robustness against sophisticated attacks.
method Introduces SOVR loss function that switches from cross-entropy to one-vs-the-rest loss for important samples.
result SOVR loss increases logit margins of important samples, improving robustness against Auto-Attack.
We prove new fast learning rates for the one-vs-all multiclass plug-in classifiers trained either from exponentially strongly mixing data or from data generated by a converging drifting distribution. These are two typical scenarios where training data are not iid. The learning rates are obtained under a multiclass vers…
New research shows SVM and related methods can overfit without harm in multiclass classification.
problem Understanding benign overfitting in multiclass classification.
method Analyzing three training algorithms: ERM with cross-entropy, least-squares, and one-vs-all SVM.
result All three algorithms can lead to classifiers that interpolate training data and have equal accuracy under high overparameterization.
Paper designs optimal ECOCs using IP for robust multiclass classification.
problem Designing robust ECOCs for multiclass classification.
method Integer Programming formulation to minimize codebooks with desirable error-correcting properties, leveraging graph-theoretic structure and edge clique covers.
result IP-generated codebooks achieve high nominal and robust adversarial accuracy.
Extends PCVM for multi-class classification with improved accuracy.
problem Lack of probabilistic outputs and contradictory predictions in multi-class classification.
method Proposes mPCVM with two learning algorithms: top-down and bottom-up.
result Superior performance, especially with many classes, validated on synthetic and benchmark data.
Survey of multi-class classification methods and their performance.
problem Generalizing binary classification to multi-class problems.
method Recursive control language to describe partitioning strategies, manual and automatic design.
result One-vs.-one is generally the best solution, but adaptive methods can outperform it in some cases.
New method selects direct causal parents from large sets of variables.
problem Inferring direct causal parents from many variables, especially nonlinear and cyclic.
method One-vs.-the-rest feature selection approach with theoretical guarantees.
result Significant improvements over existing methods.
A new meta-learning method for faster adaptive multiclass classification.
problem Efficiently designing multiclass classifiers that adaptively merge classes into a hierarchical structure.
method Adaptive hierarchical classification tree construction using empirical Bayes error rates and minimal spanning trees.
result The proposed method can learn faster and achieve competitive accuracy compared to existing methods.
Improves uncertainty estimation and OOD detection in neural networks.
problem Accurate uncertainty estimation and OOD detection in neural networks.
method Investigates one-vs-all and distance-based logit representations for probabilities.
result One-vs-all formulations improve calibration without additional complexity.
Proposes a new method to improve multiclass probability calibration.
problem Uncalibrated class probabilities in multiclass classifiers leading to over-confidence.
method Dirichlet calibration method applicable to any model class, derived from Dirichlet distributions.
result Improved probabilistic predictions across various datasets and classifiers.
Paper proposes an approximate margin method for fast multi-class classification.
problem Challenges in multi-class classification with many classes.
method Uses ANN search structures and LSH for approximate margin estimation.
result Approximate margin method is highly competitive in time, memory, and performance.
A robust multiclass SVM tackles imbalanced data uncertainty.
problem Imbalanced multiclass classification with uncertain labels.
method Wasserstein distributionally robust optimization for multiclass SVM.
result Our model outperforms state-of-the-art OVA models in imbalanced settings.
New method builds robust trees from noisy data.
problem Building accurate classification trees from noisy labeled data.
method Combines SVM-like splitting rules and label noise detection.
result Effective in detecting and mitigating label noise.
Radar-based classification improves accuracy of autonomous driving by identifying new classes.
problem Challenges in classifying sparse radar data for autonomous driving.
method Ensemble of classifiers using one-vs-one and one-vs-all strategies, with feature selection.
result Improved classification performance and identification of novel classes.
A new L2D system produces calibrated probabilities of expert correctness without sacrificing accuracy.
problem Calibration of learning to defer systems for safety.
method One-vs-all classifiers with a consistent surrogate loss function.
result Proposes a calibrated L2D system that outperforms existing methods in accuracy and calibration.
Substantial evidence indicates that major psychiatric disorders are associated with distributed neural dysconnectivity, leading to strong interest in using neuroimaging methods to accurately predict disorder status. In this work, we are specifically interested in a multivariate approach that uses features derived from …
Sparse classifiers such as the support vector machines (SVM) are efficient in test-phases because the classifier is characterized only by a subset of the samples called support vectors (SVs), and the rest of the samples (non SVs) have no influence on the classification result. However, the advantage of the sparsity has…
Bonsai learns fast, deep trees for XMC with fast training and high accuracy.
problem Efficiently learning multi-label classification models with millions of labels.
method Develops Bonsai suite of algorithms that generalize label representation and learn shallow trees.
result Bonsai achieves best of fast training and high accuracy on XMC tasks.
Adversarial approach improves extreme multi-label classification performance.
problem Learning relevant labels from datasets with many rare labels.
method Robust optimization framework with Hamming loss for tail-label detection.
result Proved efficacy of Hamming loss and improved performance over state-of-the-art methods.
Bayesian method improves few-shot classification accuracy.
problem Few-shot classification with small labeled datasets.
method Gaussian process classifier with Pólya-Gamma augmentation and one-vs-each softmax.
result Improved accuracy and uncertainty quantification.
New sEMG dataset for ADL activities recognized with high accuracy.
problem Developing a robust dataset for sEMG-based ADL activity recognition.
method Acquired sEMG data from 25 subjects performing 22 ADL activities. Used 4 classifiers with various feature sets.
result SVM classifier achieved 83.21% accuracy on 5 FAABOS categories.
Paper identifies resting positions using EGG, ECG, respiration rate, and SpO2.
problem Identifying the resting position for health monitoring.
method Hybrid stacked ensemble machine learning model combining Decision tree, Random Forest, and Xgboost.
result 100% accurate prediction of resting positions.
This paper improves multi-class calibration methods using mutual information maximization-based binning.
problem Calibration of deep neural network predictions, especially for small prior classes.
method I-Max concept for binning, shared class-wise calibration strategy.
result Improves multi-class ranking and calibration performance using a small calibration set.
New linear algorithms improve wSVMs for multiclass probability estimation.
problem Estimating conditional probabilities for multiclass problems.
method Proposed baseline learning and OVA learning schemes to improve wSVMs.
result Linear algorithms achieve optimal computational efficiency and good estimation accuracy.
Proposed SMO algorithm for OC-SVM+ significantly outperforms non-sequential algorithms.
problem One-class SVM with privileged information
method Sequential Minimal Optimization (SMO) algorithm
result Finite-time convergence established
Support Vector Machines, SVMs, and the Large Margin Nearest Neighbor algorithm, LMNN, are two very popular learning algorithms with quite different learning biases. In this paper we bring them into a unified view and show that they have a much stronger relation than what is commonly thought. We analyze SVMs from a metr…
Unified Pin-SVM improves accuracy over existing Pin-SVM model.
problem Difficulty in Pin-SVM model for −1≤τ<0. method Unified Pin-SVM model that solves a QPP for −1≤τ≤1. result Significant improvement in accuracy over existing Pin-SVM model.
Localized SVMs maintain SVM's consistency properties for large datasets.
problem Inefficient computational requirements of global SVMs for large data sets.
method Localized SVMs apply different hyperparameters to different regions of the input space.
result Localized SVMs inherit Lp- and risk consistency from global SVMs. MU-SVM improves multiclass classification accuracy.
problem Multiclass classification problems.
method Proposes MU-SVM for multiclass learning and an analytic span bound for model selection.
result Achieves > 20% improvement in test accuracies compared to multi-class SVM.
This paper analyzes the impact of asynchronous updates on OVA models' accuracy.
problem The impact of asynchronous updates on the accuracy of OVA models.
method Defined a metric to quantify dataset differences, analyzed three factors (number of classes, data points, and training dataset divergence), and evaluated Spoken Language Understanding system.
result The proposed metric correlates strongly with model performances.
BAEN-SVM improves SVM robustness to noisy data.
problem Noise and geometric irrationalities in SVM.
method Bounded asymmetric elastic net loss combined with SVM.
result BAEN-SVM is robust to noise and geometrically well-defined.
This work extends SVM error bounds to weighted SVM and introduces hyperparameter selection methods.
problem Improving SVM performance through effective hyperparameter selection.
method Extending span error bound theory to weighted SVM and introducing hyperparameter selection methods.
result The span rule is the most effective method for weighted SVM hyperparameter selection and provides the best predictor of test error.
New SVM feature selection methods improve wafer testing accuracy.
problem Need for improved SVM feature selection methods.
method Entropy measurement and K-medoid clustering for feature selection.
result One-class SVM's enhance wafer testing accuracy.
Support vector machines (SVMs) are invaluable tools for many practical applications in artificial intelligence, e.g., classification and event recognition. However, popular SVM solvers are not sufficiently efficient for applications with a great deal of samples as well as a large number of features. In this paper, thus…
New method reduces labeler costs by aggregating predictions from local classifiers.
problem Reduce labeler costs in multiclass classification.
method Model K-class classification using smaller classifiers trained on subsets of tasks. result Near-optimal scheme for designing classifier configurations reduces labeler costs.
A quantum-inspired classical algorithm speeds up LS-SVM classification.
problem Big data challenge in SVM classification.
method Improved indirect sampling technique for LS-SVM.
result Algorithm achieves logarithmic runtime for low rank data matrices.
This paper improves SVM prediction uncertainty quantification methods.
problem Lack of comprehensive UQ methods for SVM predictions.
method Developed SSVQR model for sparse PI estimation and feature selection algorithm.
result Proposed SSVQR model achieves sparse solutions and improves PI quality.
Introduces Soft-SVM for binary classification bridging logistic and SVM.
problem Data separability issues in binary classification.
method Soft-SVM regression using convex relaxation of hinge loss with softness and class-separation parameters.
result Soft-SVM performs well in classification and prediction errors.
Paper addresses SVM bias in high-dimension, low-sample-size settings.
problem Bias in SVM performance in high-dimension, low-sample-size settings.
method Proposes a bias-corrected SVM (BC-SVM) to improve SVM performance.
result BC-SVM gives preferable performances in high-dimension, low-sample-size settings.
Paper proposes an ensemble SVM method for efficient VAD.
problem Efficient and accurate VAD for speech processing.
method Supervised learning with ensemble SVM on large datasets.
result Ensemble SVM outperforms stand-alone SVM in VAD accuracy.
GADGET SVM uses gossip-based distributed learning for scalable SVMs.
problem Scalability issues in traditional SVM algorithms for large datasets.
method Gossip-based distributed learning for the primal SVM formulation.
result Performance comparable to centralized and online SVM algorithms.
Proposes SVM-based Deep Stacking Network for improved deep learning.
problem Improving deep learning performance and interpretability.
method Uses stacked SVM classifiers within a DSN architecture and a BP-like layer tuning scheme.
result Demonstrates superior performance compared to benchmark models on image and text data.
Prior knowledge can be used to improve predictive performance of learning algorithms or reduce the amount of data required for training. The same goal is pursued within the learning using privileged information paradigm which was recently introduced by Vapnik et al. and is aimed at utilizing additional information avai…
This paper improves SVM classification using a differentiable loss function and a gradient method.
problem Improving SVM classification with a differentiable loss function.
method Uses the Huberized Support Vector Machine (HSVM) and Proximal Gradient (PG) method.
result The proposed method converges linearly and supports the solution in finite time.
Abstract: Bounds for multi-task and multi-category learning.
problem Risk bounds for vector-valued functions.
method Framework for deriving risk bounds with feature maps and loss functions.
result Conditions for shared representations beneficial in multi-task and multi-category learning.
BrainSurfCNN predicts task contrasts from resting-state fingerprints, improving accuracy over baseline.
problem Predicting task-evoked activity from resting-state functional connectivity.
method Surface-based convolutional neural network (BrainSurfCNN) with reconstructive-contrastive loss.
result Significantly improved accuracy in predicting task contrasts over baseline.
L3-SVMs clusters data, reduces dimensions, and learns linear models.
problem Capturing non-linearities and scaling to large datasets.
method Clusters input space, projects data onto landmarks, learns linear combination of local models.
result L3-SVMs achieves generalization guarantees and competitive performance.