This paper considers the challenge of evaluating a set of classifiers, as done in shared task evaluations like the KDD Cup or NIST TREC, without expert labels. While expert labels provide the traditional cornerstone for evaluating statistical learners, limited or expensive access to experts represents a practical bottl…
On-device research index
arXiv research
A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
168,657 papers · 148 categories
Trend · papers per month
48 results for “Classifier Evaluation”
New IRT method identifies useful datasets for ML classifier evaluation.
problem Lack of standard evaluation strategy for ML benchmarks.
method Applied Item Response Theory (IRT) to OpenML-CC18 benchmark.
result Not all datasets are useful for evaluating classifiers.
Paper develops streaming algorithms to estimate classifier accuracy on unlabeled data.
problem Estimating classifier accuracy on unlabeled data with noisy decisions.
method Two algebraic evaluators: majority voting and a novel method to handle correlated classifiers.
result The novel method can be as accurate as 1% when handling small amounts of correlation.
Paper introduces a method to evaluate abstaining classifiers by considering missing predictions as counterfactuals.
problem Lack of a principled approach to evaluate and compare abstaining classifiers, especially when missing predictions are important.
method Develops a novel approach to treat abstentions as missing data, defining counterfactual scores and using causal inference methods to estimate them.
result Shows that under certain conditions, counterfactual scores can be identified and estimated efficiently, improving evaluation of abstaining classifiers.
We introduce a new metric to evaluate corruption robustness of ML classifiers.
problem Evaluating corruption robustness of machine learning classifiers.
method We propose a test data augmentation method using minimal class separation distance to derive a robustness distance ε and a metric MSCR.
result The MSCR metric allows interpretable comparison of classifier robustness on different datasets.
The use of alternative measures to evaluate classifier performance is gaining attention, specially for imbalanced problems. However, the use of these measures in the classifier design process is still unsolved. In this work we propose a classifier designed specifically to optimize one of these alternative measures, nam…
Proposes a method to evaluate classifiers with missing labels using multiple imputation.
problem Missing labels during model evaluation can introduce bias, especially in Missing Not At Random (MNAR) data.
method Develops a multiple imputation technique to estimate and provide predictive distributions for metrics like precision, recall, and ROC-AUC.
result The predictive distribution's location and shape are generally correct, even in the MNAR regime.
New neural network approach using mutual information.
problem Training neural networks for imbalanced datasets.
method Converts neural network classifiers to mutual information evaluators.
result New form of softmax leads to better classification accuracy, especially for imbalanced datasets.
EAST aligns neural network classifiers with user-defined evaluation metrics.
problem Mismatch between neural network training and evaluation metrics leads to suboptimal performance.
method EAST uses dynamic thresholding, soft-set confusion matrix, and annealing to align neural network predictions with target evaluation metrics.
result EAST improves alignment between training objectives and evaluation metrics, outperforming existing methods.
This paper reviews PU learning evaluation methods and provides practical recommendations.
problem Evaluating PU learning methods when only positive and unlabelled data are available.
method Critical review of 51 articles proposing PU classifiers and alternative predictive accuracy measures.
result Practical recommendations for improving PU learning evaluation.
This work addresses fairness in ML models by training and evaluating attribute classifiers under uncertain and incomplete data.
problem Challenges in fairness metrics due to uncertain and incomplete data.
method Developed a theoretical and empirical analysis to understand and improve bias estimation in the data-scarce regime.
result The test accuracy of the attribute classifier is not always correlated with its effectiveness in bias estimation.
This work introduces significativity indices for agreement values between classifiers.
problem Evaluating the quality of agreement measures between classifiers.
method Proposes general approach and two specific indices for evaluation.
result Introduces two new significativity indices for evaluating agreement values.
Improves classifier evaluation by aligning with Total Classification Cost.
problem Lack of consensus on evaluation metrics and class imbalance issues.
method Introduces Weighted Accuracy (WA) and a reweighting framework for cost-sensitive scenarios.
result WA aligns with Total Classification Cost (TCC) minimization under realistic conditions.
New robustness metric helps select reliable classifiers.
problem Evaluating reliability of classifier predictions.
method Proposed new robustness metric for any classifier and feature type.
result Demonstrated ability to distinguish reliable from unreliable predictions.
Probabilistic classifiers output a probability distribution on target classes rather than just a class prediction. Besides providing a clear separation of prediction and decision making, the main advantage of probabilistic models is their ability to represent uncertainty about predictions. In safety-critical applicatio…
Proposes MCC-F1 curve for better binary classification evaluation.
problem Misleading performance evaluations with ROC and PR curves for imbalanced data.
method Introduces MCC-F1 curve combining MCC and F1 score.
result MCC-F1 curve provides clearer classifier differentiation.
The paper extends calibration to sets of probabilistic classifiers, finding many ensembles are poorly calibrated.
problem Evaluating the validity of epistemic uncertainty in sets of probabilistic classifiers.
method Proposed a novel nonparametric calibration test for sets of probabilistic classifiers.
result Ensembles of deep neural networks are often not well calibrated.
Defines explanations for classifier outcomes using causal concepts.
problem Understanding classifier outcomes in a causal context.
method Proposes a new definition of explanation based on causality, compares it with existing notions, and evaluates it experimentally.
result Experimental evaluation shows the new definition's effectiveness on financial datasets.
Enhances mixture models with classifier-defined weights.
problem Density evaluation and sampling in mixture models.
method Introduces Classifier Weighted Mixtures (CWM) with functional weights.
result Improves expressivity in variational estimation without increasing complexity.
Temporal cross-validation impacts multivariate time series subsequence anomaly detection evaluationstat.ML
Study examines how different time series cross-validation methods affect anomaly detection in multivariate time series.
problem Evaluating anomaly detection in multivariate time series requires preserving temporal dependencies, especially for subsequence anomalies.
method Systematically investigates walk-forward and sliding window methods across various validation configurations and classifier types.
result Sliding window method consistently yields higher precision-recall scores and reduced fold-to-fold performance variance, particularly for deep learning models.
Revisit Time Series Classification Benchmark: The Impact of Temporal Information for Classificationcs.LG
Temporal information impacts only a fraction of time series datasets, skewing benchmark evaluations.
problem Temporal information's impact on time series classification is often overestimated.
method Permutation tests on UCR archive to identify datasets where temporal info is irrelevant.
result Many tabular datasets perform well without temporal info, skewing benchmark evaluations.
PolyGraph Discrepancy improves graph generative model evaluation.
problem Inability of existing metrics to provide an absolute performance measure and comparability across different graph descriptors.
method Approximates Jensen-Shannon distance using binary classifiers trained to distinguish between real and generated graphs.
result PGD provides a more robust and insightful evaluation compared to MMD metrics.
We present a method to stop the evaluation of a prediction process when the result of the full evaluation is obvious. This trait is highly desirable in prediction tasks where a predictor evaluates all its features for every example in large datasets. We observe that some examples are easier to classify than others, a p…
We present a method to stop the evaluation of a decision making process when the result of the full evaluation is obvious. This trait is highly desirable for online margin-based machine learning algorithms where a classifier traditionally evaluates all the features for every example. We observe that some examples are e…
Generative text classifiers are most vulnerable to membership inference attacks.
problem Privacy threat from Membership Inference Attacks on generative text classifiers.
method Comprehensive empirical evaluation of generative, discriminative, and pseudo-generative classifiers across various datasets.
result Generative classifiers explicitly modeling are most vulnerable to membership leakage.
Motivated by safety-critical classification problems, we investigate adversarial attacks against cost-sensitive classifiers. We use current state-of-the-art adversarially-resistant neural network classifiers [1] as the underlying models. Cost-sensitive predictions are then achieved via a final processing step in the fe…
Unsupervised classification methods learn a discriminative classifier from unlabeled data, which has been proven to be an effective way of simultaneously clustering the data and training a classifier from the data. Various unsupervised classification methods obtain appealing results by the classifiers learned in an uns…
Unified approach optimizes neural network training for various metrics.
problem Training and evaluation of neural network binary classifiers often use different metrics.
method Combines differentiable approximation and probabilistic soft sets.
result Effective in optimizing for metrics like F1-Score across various domains.
Given a classifier ensemble and a set of examples to be classified, many examples may be confidently and accurately classified after only a subset of the base models in the ensemble are evaluated. This can reduce both mean latency and CPU while maintaining the high accuracy of the original ensemble. To achieve such gai…
This paper introduces metrics to evaluate robustness of neural networks to natural adversarial examples.
problem Measuring robustness of neural networks to natural adversarial examples.
method Proposes latent space performance metrics based on generative models.
result Latent adversarial perturbations are often perceptually small and associated with classifier accuracy.
Study improves image classifier robustness to random p-norm corruptions.
problem Improving robustness of image classifiers to real-world imperceptible corruptions.
method Training and testing with random p-norm corruptions, evaluating robustness against different p-norms.
result Training with a combination of p-norm corruptions significantly improves robustness.
We explore adversarial robustness in the setting in which it is acceptable for a classifier to abstain---that is, output no class---on adversarial examples. Adversarial examples are small perturbations of normal inputs to a classifier that cause the classifier to give incorrect output; they present security and safety …
Proposes and evaluates three diagnostic graphics for probabilistic classifiers.
problem Evaluating and comparing probabilistic classifiers.
method Triptych of diagnostic graphics: reliability diagram, ROC curve, Murphy diagram.
result Visual diagnostics reveal distinct aspects of forecast performance.
Estimation of importance sampling weights for off-policy evaluation of contextual bandits often results in imbalance - a mismatch between the desired and the actual distribution of state-action pairs after weighting. In this work we present balanced off-policy evaluation (B-OPE), a generic method for estimating weights…
Suitability filter detects model performance degradation in real-world deployment.
problem Ensuring model reliability in safety-critical domains without access to ground truth labels.
method Uses suitability signals to evaluate classifier performance on unlabeled user data.
result The suitability filter reliably detects performance deviations due to covariate shift.
The study analyzes performance indices for class-imbalanced data and identifies conditions they must meet.
problem Distortions in performance indices under class imbalance.
method Identified two conditions for performance indices and analyzed four binary and five multi-class indices.
result Recommended appropriate indices for evaluating classifiers in class-imbalanced scenarios.
In this paper we establish rigorous benchmarks for image classifier robustness. Our first benchmark, ImageNet-C, standardizes and expands the corruption robustness topic, while showing which classifiers are preferable in safety-critical applications. Unlike recent robustness research, this benchmark evaluates performan…
Improves multi-label classification with a new network model.
problem Improving multi-label classification accuracy.
method Introduces Classifier Chain Network (CCN) for multi-label classification.
result CCN outperforms benchmark methods in simulations and real data.
In Dynamic Ensemble Selection (DES) techniques, only the most competent classifiers are selected to classify a given query sample. Hence, the key issue in DES is how to estimate the competence of each classifier in a pool to select the most competent ones. In order to deal with this issue, we proposed a novel dynamic e…
Generative models are known to be difficult to assess. Recent works, especially on generative adversarial networks (GANs), produce good visual samples of varied categories of images. However, the validation of their quality is still difficult to define and there is no existing agreement on the best evaluation process. …
New framework for evaluating multiclass classifier calibration.
problem Ensuring classifiers are well-calibrated for trustworthy predictions.
method Utility Calibration framework that measures calibration error relative to a utility function.
result Unified and robust interpretation of existing calibration metrics.
SPOCC: Scalable POssibilistic Classifier Combination -- toward robust aggregation of classifierscs.LG
We investigate a problem in which each member of a group of learners is trained separately to solve the same classification task. Each learner has access to a training dataset (possibly with overlap across learners) but each trained classifier can be evaluated on a validation dataset. We propose a new approach to aggre…
Image classifiers often suffer from adversarial examples, which are generated by strategically adding a small amount of noise to input images to trick classifiers into misclassification. Over the years, many defense mechanisms have been proposed, and different researchers have made seemingly contradictory claims on the…
This work proposes optimal decision rules for hierarchical classifiers to better align with evaluation metrics.
problem Heuristic decision rules in hierarchical classification do not align with evaluation metrics.
method Derives optimal decision rules for various prediction settings, focusing on hierarchical scores.
result Optimal decision rules enhance the performance and reliability of hierarchical classifiers.
New method to evaluate visual explanations from neural networks.
problem Lack of consensus on measuring effectiveness of visual explanations.
method Proposed a new procedure for evaluating explanations using a range of sources.
result Demonstrated the benefit of combining different sources and the impact of bias parameters.
Image classification has become one of the main tasks in the field of computer vision technologies. In this context, a recent algorithm called CapsNet that implements an approach based on activity vectors and dynamic routing between capsules may overcome some of the limitations of the current state of the art artificia…
Deep neural networks (DNN) are able to successfully process and classify speech utterances. However, understanding the reason behind a classification by DNN is difficult. One such debugging method used with image classification DNNs is activation maximization, which generates example-images that are classified as one o…
In this paper, we empirically evaluate algorithms for learning four types of Bayesian network (BN) classifiers - Naive-Bayes, tree augmented Naive-Bayes, BN augmented Naive-Bayes and general BNs, where the latter two are learned using two variants of a conditional-independence (CI) based BN-learning algorithm. Experime…