arXiv research

Learnable multiclass hypothesis classes don't always have a sample compression scheme of fixed size.

This work characterizes when a hypothesis class can be k-list learned.

New research determines the optimal sample complexity for multiclass and list learning.

New algorithm improves multiclass classification regret bound.

Study optimal rates for multiclass classification, resolving open questions.

Study expands multiclass classification models with new rates and partial concept classes.

New Sauer inequality improves multiclass hypothesis class bounds.

Characterizes learnability of forgiving 0-1 loss functions in multiclass settings.

Local regularization fails in transductive learning for some multiclass problems.

Proper learning is possible with labeled data, but unlabeled data can improve performance.

Study robust online learning with adversarial perturbations.

Enhanced metrics for multiclass classification improve on existing methods.

Study shows SQ hardness for multiclass linear classification with random noise.

Improved PAC learning algorithm for multiclass classification with bandit feedback.

The paper analyzes how overparameterized models can generalize well in multiclass classification.

New framework for evaluating multiclass classifier calibration.

Research on predicting with lists of labels, characterizing learnability and providing algorithms.

Pre-trained word embeddings encode general word semantics and lexical regularities of natural language, and have proven useful across many NLP tasks, including word sense disambiguation, machine translation, and sentiment analysis, to name a few. In supervised tasks such as multiclass text classification (the focus of …

Improved multiclass classification with class-weighted nearest neighbors.

Novel method for multiclass ROC curves using multidimensional Gini index.

Proposes MCLLO for assessing and recalibrating multiclass probability predictions.

We consider a model of robust learning in an adversarial environment. The learner gets uncorrupted training data with access to possible corruptions that may be affected by the adversary during testing. The learner's goal is to build a robust classifier, which will be tested on future adversarial examples. The adversar…

Proposes top-label calibration and M2B framework for multiclass to binary calibration.

New algorithm learns multiclass concepts with finite Littlestone dimension.

Algorithm minimizes risk for multiclass classification of stochastic diffusion paths.

Generalized additive models (GAMs) are favored in many regression and binary classification problems because they are able to fit complex, nonlinear functions while still remaining interpretable. In the first part of this paper, we generalize a state-of-the-art GAM learning algorithm based on boosted trees to the multi…

New binary classification techniques help multiclass classification by aggregating proper learners.

We introduce Universum learning for multiclass problems and propose a novel formulation for multiclass universum SVM (MU-SVM). We also propose an analytic span bound for model selection with almost 2-4x faster computation times than standard resampling techniques. We empirically demonstrate the efficacy of the proposed…

Unified binary and multiclass margin-based classification methods.

Multiclass prediction is the problem of classifying an object into a relevant target class. We consider the problem of learning a multiclass predictor that uses only few features, and in particular, the number of used features should increase sub-linearly with the number of possible classes. This implies that features …

We present a graph-based variational algorithm for multiclass classification of high-dimensional data, motivated by total variation techniques. The energy functional is based on a diffuse interface model with a periodic potential. We augment the model by introducing an alternative measure of smoothness that preserves s…

Class ambiguity is typical in image classification problems with a large number of classes. When classes are difficult to discriminate, it makes sense to allow k guesses and evaluate classifiers based on the top-k error instead of the standard zero-one loss. We propose top-k multiclass SVM as a direct method to optimiz…

Paper analyzes multiclass classification with high-dimensional data.

Develops NPMC method for noisy labels, improving multiclass classification accuracy.

New method for multiclass classification reduces error bounds.

Unified SVM framework tackles multiclass and multilabel classification.

FOLKLORE algorithm speeds up online multiclass logistic regression.

Conventional multiclass conditional probability estimation methods, such as Fisher's discriminate analysis and logistic regression, often require restrictive distributional model assumption. In this paper, a model-free estimation method is proposed to estimate multiclass conditional probability through a series of cond…

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…

Study extends binary omniprediction to multiclass setting with improved sample complexity.

This work extends implicit bias analysis to multiclass classification using a new loss framework.

A new method learns proper multiclass losses and probabilities.

Study builds a classifier for diffusions with unknown diffusion but known drifts.

We develop efficient algorithms to train $\ell_1$-regularized linear classifiers with large dimensionality $d$ of the feature space, number of classes $k$, and sample size $n$. Our focus is on a special class of losses that includes, in particular, the multiclass hinge and logistic losses. Our approach combines several…

New reformulations for multiclass classification problems using optimal transport.

New linear algorithms improve wSVMs for multiclass probability estimation.

A new KDE-based method improves multiclass quantification.

Distillation improves simple models by approximating complex labels.