Proposes an angle-based framework for multicategory cost-sensitive classification.
problem Cost-sensitive multicategory classification challenges.
method Angle-based cost-sensitive classification framework without sum-to-zero constraint.
result Proposed boosting algorithms yield competitive classification performances.
New refit strategy improves probability estimation for multicategory angle-based classifiers.
problem Improving probability estimation for multicategory angle-based classifiers in high dimensional applications.
method Proposes a new refit strategy for multicategory angle-based classifiers, adding small computation cost.
result Significant improvement in probability estimation with minimal additional computation.
Predict steering angles of self-driving cars from images.
problem Predicting steering angles for self-driving cars using image data.
method Used deep learning techniques like Transfer Learning, 3D CNN, LSTM, and ResNet to predict steering angles.
result Both models placed in the top ten of Udacity's challenge.
Develops a novel approach for estimating optimal DTRs with multicategory treatments and censored data.
problem Estimating optimal treatment regimes for chronic diseases with censored data.
method Angle-based multicategory classification algorithm for maximizing conditional survival function.
result The proposed method outperforms existing approaches in maximizing conditional survival function.
Interpretable ML models for missing data and visualisation.
problem Understanding and evaluating fairness in ML models.
method Introduced angle-based variants of Learning Vector Quantization (LVQ) models.
result Models can handle missing values and extract knowledge from datasets.
We compute, using a formula of Dittmann, the Bures metric tensor (g) for the eight-dimensional convex set of three-level quantum systems, employing a newly-developed Euler angle-based parameterization of the 3 x 3 density matrices. Most of the individual metric elements (g_{ij}) are found to be expressible in relativel…
New method estimates intrinsic dimensionality using angles, not distances.
problem Estimating local intrinsic dimensionality accurately.
method Introduces a new estimator using the distribution of angles between neighbor points.
result New estimator behaves similarly but complementarily to existing measures of intrinsic dimensionality.
AJIVE captures joint and individual variation in diverse data blocks.
problem Integrating multiple genomic data sets to understand cancer genetics and cell biology.
method Angle-Based Joint and Individual Variation Explained (AJIVE) using score subspaces and perturbation theory.
result AJIVE provides a fast, linear algebra computation method insensitive to data heterogeneity.
Curvature regularization prevents distortion in graph embeddings.
problem Graph topology patterns distort in Euclidean space, making detection difficult.
method Proposes curvature regularization to enforce flatness in embedding manifolds.
result Significant improvements in five embedding methods on open graph datasets.
Unified framework for optimizing treatment quantiles considering risk and efficacy.
problem Maximizing treatment outcomes while controlling risk in sequential clinical decisions.
method Risk-Aware Quantile Dynamic Treatment Regimes (RQDTR) framework that optimizes a prespecified quantile of the cumulative potential outcome while incorporating treatment-related risk.
result Improves tail-oriented efficacy and achieves more favorable benefit-risk trade-offs compared to existing methods.
Proposes HeteroJIVE for joint subspace estimation in multi-view data with statistical and structural heterogeneity.
problem Joint subspace estimation in multi-view data with varying statistical and structural heterogeneity.
method HeteroJIVE: A weighted two-stage spectral algorithm addressing statistical and structural heterogeneity.
result HeteroJIVE achieves the O(K−1/2) rate without iterative refinement, validating the oracle-optimal weighting scheme. LOCUS separates brain network connectivity matrices efficiently.
problem High dimensionality, latent sources, and spurious findings in analyzing brain connectivity matrices.
method LOCUS: low-rank structure with uniform sparsity, iterative Node-Rotation algorithm.
result LOCUS achieves more efficient and accurate source separation for connectivity matrices.
This paper analyzes AJIVE for estimating shared subspace across multiple datasets, revealing its strengths and limitations.
problem Estimating shared subspace across multiple datasets with varying degrees of misalignment.
method Angle-based Joint and Individual Variation Explained (AJIVE) method, a two-stage spectral approach.
result AJIVE's performance in high signal-to-noise ratio (SNR) regimes and its non-diminishing error in low-SNR settings.
The paper explains how ReLU nets converge globally in high dimensions without strict assumptions.
problem Understanding global convergence of ReLU nets in very high dimensions.
method Fine-grained analysis of random activation matrices and detailed gradient norm and curvature analysis.
result Empirical loss function has favorable geometrical properties in the overparameterized setting.
Paper tackles unsupervised domain adaptation for unlabeled target domains.
problem Existing domain adaptation research focuses on labeled target domains, ignoring unlabeled ones.
method Developed an unsupervised knowledge transfer theorem and metric, implemented in GLG model.
result Proposed model outperforms existing baselines across three applications.
New method for robust financial portfolio analysis.
problem Challenges in modeling financial portfolio dependence structure.
method Nonparametric Angles-based Correlation (NAbC) method.
result Valid inferences and flexible scenarios for portfolio analysis.
Bayesian network classifiers are used in many fields, and one common class of classifiers are naive Bayes classifiers. In this paper, we introduce an approach for reasoning about Bayesian network classifiers in which we explicitly convert them into Ordered Decision Diagrams (ODDs), which are then used to reason about t…
A new trust score improves classifier reliability.
problem Improving classifier reliability in safety-critical applications.
method Proposed a trust score based on classifier agreement with a modified nearest-neighbor classifier.
result Empirically shows high trust scores produce high precision in identifying correctly classified examples.
Study on the structure of classifier boundaries in DNA sequencing.
problem Understanding the structure of boundaries in a Bayes classifier for DNA sequencing.
method Examined the structure of the boundary in a Bayes classifier applied to DNA sequencing data. Introduced a new measure of uncertainty, Neighbor Similarity.
result The boundary is large and complex, and Neighbor Similarity effectively measures classifier uncertainty.
RCAM-based ensemble combines binary classifiers using similarity and vote scheme.
problem Improving binary classification accuracy through ensemble methods.
method RCAM-based ensemble combining classifiers using similarity and recurrent consult-vote scheme.
result RCAM-based ensemble outperforms individual classifiers and majority voting.
Proposes SDWEC for improved classification accuracy and reduced classifier count.
problem Improves classification accuracy and minimizes the number of classifiers.
method Formed an ensemble of pre-trained classifiers with assigned weights, modeled as a non-convex cost function with terms for data fidelity, sparsity, and non-negativity constraints. Employed convex relaxation techniques and approximations for efficient solution.
result SDWEC provides better or similar accuracy levels using fewer classifiers and reduces testing time.
Random classifier ensembles improve fairness and diversity in decision making.
problem Loss of diversity and fairness in single machine learning classifiers.
method Study of random classifier ensembles for fairness-aware learning.
result Ensemble of classifiers can achieve better fairness and accuracy trade-offs.
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…
Boosting classifiers improve accuracy with noisy inputs.
problem Noisy communication or computation degrades boosting classifier accuracy.
method Optimize resource allocation for base classifiers based on importance metrics.
result Optimized noisy boosting classifiers are more robust than bagging.
ECPF improves classification accuracy and speed for evolving data streams.
problem Reusing classifiers trained on recurring concepts to maintain accuracy and speed.
method ECPF uses similarity of classifications to quickly identify the best classifier to reuse.
result ECPF significantly outperforms state-of-the-art frameworks on synthetic and real-world datasets.
New method computes discriminative classifiers from generative models.
problem Discriminative vs generative classifiers are often seen as distinct, but this work shows they can be equivalent.
method General theoretical result showing generative classifiers can be computed discriminatively.
result Bayesian Maximum Posterior classifier from generative models matches discriminative classifier definition.
Generative classifier derived from any discriminative classifier rejects illegal inputs.
problem Detecting and rejecting illegal inputs like adversarial examples and out-of-distribution samples.
method SDIM-logit: learns generative classifier from logits of any discriminative classifier, imposing statistical constraints.
result SDIM-logit inherits performance of base classifier without loss and can reject illegal inputs.
Simple construction for classifying spaces of projections of immersions.
problem Classifying spaces for projections of immersions with controlled singularities.
method Explicit simple construction for classifying spaces of maps obtained as hyperplane projections of immersions.
result Structure theorems for these classifying spaces.
Study examines how classifier performance is affected by training data quality.
problem How classifier performance is affected by training data quality.
method Extensive numerical experiments with four classifiers (Bayes, neural nets, partition models, random forests) on metagenomic assembly data.
result Classifier performance degrades as training data quality degrades, leading to breakdown-like behavior.
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.
FOCA method prevents co-adaptation between feature extractor and classifier.
problem Co-adaptation between feature extractor and classifier degrades neural network performance.
method FOCA method uses randomly-generated, weak classifiers to optimize feature extractor without explicit co-adaptation.
result FOCA features form a point-like distribution within the same class under special conditions.
A novel KD classifier improves character recognition performance.
problem Improving character recognition accuracy.
method Kernel-based generative classifier in distortion subspace with iterative kernel selection.
result The KD classifier outperforms existing classifiers and has unique recognition capability.
A review of statistical SSL methods showing improved classifier performance.
problem Forming classifiers from limited labeled data and many unlabeled data.
method Statistical approaches to semi-supervised learning.
result A classifier from partially labeled data can have lower expected error rate.
Paper proposes a classifier that optimizes utility function with prior knowledge.
problem Designing a classifier that optimizes a utility function based on prior knowledge.
method Systematic framework incorporating prior knowledge to optimize a utility function.
result The classifier asymptotically converges to the optimal classifier (Bayes rule) as data size grows.
Deep Bayes classifiers are more robust to adversarial attacks than discriminative classifiers.
problem Robustness of deep neural network classifiers against adversarial attacks.
method Developed deep Bayes classifier using conditional deep generative models and detection methods.
result Deep Bayes classifiers are more robust than deep discriminative classifiers.
A bias classifier is introduced to resist adversarial attacks.
problem Resisting adversarial attacks on deep neural networks (DNNs).
method Introducing the bias part of a DNN with Relu as the activation function as a classifier, and adding a random first-degree part to make it information-theoretically safe.
result The bias classifier is more robust than DNNs of similar size against adversarial attacks.
New method controls classifier guidance in diffusion models.
problem Improving classifier guidance in diffusion models.
method Cross-entropy control of classifier gradients.
result Effective guidance vectors with mean squared error O(dε). Paper introduces algorithms for explaining monotonic classifiers.
problem Need for explanations of monotonic classifiers.
method Polynomial algorithms for formal explanations of monotonic classifiers.
result Efficient model-agnostic algorithm for enumerating explanations.
This paper evaluates criteria for a versatile classifier.
problem Choosing an optimal classifier for all problems.
method Data analysis and scoring of six popular classifiers.
result Random forests are the best generalist classifiers.
Generative classifiers' properties are linked to linear constraints.
problem Understanding the Markov property in generative classifiers.
method Characterization of discrimination functions using linear constraints and a second order finite difference operator.
result Discrimination functions of undirected Markov network classifiers are characterized by sets of linear constraints.
Ensemble validation shows selectivity penalties but variety benefits.
problem Selecting classifiers for ensemble models and their error bounds.
method Forming an ensemble from a set of hypothesis classifiers, selecting randomly, with an error bound formula.
result No penalty for using a richer hypothesis set if same fraction selected.
Ensemble quantile classifier improves performance on high-dimensional data.
problem Discriminating high-dimensional data with heavy-tailed or skewed inputs.
method Regularized quantile classifier that assigns variable weights.
result Consistently estimates minimal population loss and is Bayes optimal.
In this paper we present a new Bayesian network model for classification that combines the naive-Bayes (NB) classifier and the finite-mixture (FM) classifier. The resulting classifier aims at relaxing the strong assumptions on which the two component models are based, in an attempt to improve on their classification pe…
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 P(X,Y) are most vulnerable to membership leakage. New uniqueness concept for adversarial Bayes classifier.
problem Understanding adversarial Bayes classifiers in binary classification.
method Developed a new notion of uniqueness and analyzed it for a family of one-dimensional data distributions.
result Improved regularity of adversarial Bayes classifiers as perturbation radius increases.
A new method combines classifiers using possibility distributions and adaptive t-norms.
problem Aggregating predictions from multiple classifiers trained on overlapping datasets.
method Proposes a new approach to aggregate classifier predictions using possibility theory and adaptive t-norms.
result Proves the proposed approach possesses desirable robustness properties.
META-DES.H selects competent classifiers using meta-learning and dynamic weighting.
problem Selecting competent classifiers in dynamic ensemble selection.
method META-DES framework using meta-learning and dynamic weighting.
result Improvements in recognition accuracy on 30 datasets.
A new classifier for high-dimensional data with spiked eigenvalues.
problem Classifying high-dimensional data with spiked eigenvalues.
method Distance-based classifier using data transformation and noise reduction.
result The new classifier performs better than existing methods on simulated and real data.