New research shows the maximum ℓ1-margin classifier doesn't adapt to sparse ground truths.
problem Understanding the limitations of the maximum ℓ1-margin classifier in high-dimensional settings.
method Analyzing convergence and prediction error rates of the maximum ℓ1-margin classifier.
result Proves tight upper and lower bounds for prediction error, showing benign overfitting.
The paper analyzes boosting and minimum-ℓ1-norm classifiers in high dimensions.
problem Understanding the generalization error and optimal Bayes error in boosting.
method High-dimensional asymptotic theory, Gaussian comparison techniques, uniform deviation argument.
result Precise characterizations of boosting test error and optimal Bayes error.
AdaBoost improves binary classification in robust one-bit compressed sensing with adversarial errors.
problem Binary classification in robust one-bit compressed sensing with adversarial errors.
method AdaBoost and max-ℓ1-margin-classifier approach, with convergence rates improved under certain feature conditions. result Improved convergence rates and explanation for harmless interpolating adversarial noise.
The study classifies graphs with specific curvature and maximum degree.
problem Graphs with nonnegative Ricci curvature and maximum degree constraints.
method Classification of graphs with Lin-Lu-Yau-Ollivier Ricci curvature, maximum degree ≤ 3, and diameter ≥ 6.
result Classification of graphs meeting the specified criteria.
In many real-world applications, data is not collected as one batch, but sequentially over time, and often it is not possible or desirable to wait until the data is completely gathered before analyzing it. Thus, we propose a framework to sequentially update a maximum margin classifier by taking advantage of the Maximum…
Study shows how over-parameterized classifiers can still perform well on noisy data.
problem Understanding how maximum margin classifiers perform in over-parameterized settings with noisy data.
method Analyzes maximum margin classifiers on sub-Gaussian mixtures, providing risk bounds.
result Characterizes conditions for 'benign overfitting' in linear classification problems.
Ensemble learning is a powerful approach to construct a strong learner from multiple base learners. The most popular way to aggregate an ensemble of classifiers is majority voting, which assigns a sample to the class that most base classifiers vote for. However, improved performance can be obtained by assigning weights…
Paper proposes a method to estimate true positive proportion without knowing it.
problem Bias in binary classifier performance due to different positive item proportions.
method Maximum likelihood estimator for true proportion of positives.
result Method accurately estimates true positive proportion in data sets.
Paper develops MRCs for supervised classification using generalized maximum entropy.
problem Developing robust classifiers for decision problems.
method Generalized maximum entropy principle applied to minimax risk classifiers.
result Learning techniques for determining MRCs with performance guarantees.
Classifiers based on probabilistic graphical models are very effective. In continuous domains, maximum likelihood is usually used to assess the predictions of those classifiers. When data is scarce, this can easily lead to overfitting. In any probabilistic setting, Bayesian averaging (BA) provides theoretically optimal…
Study examines liquidation, leverage, and optimal margin requirements in Bitcoin futures markets.
problem Understanding and optimizing margin requirements in Bitcoin futures markets.
method Empirical analysis using generalized extreme value theory and BitMEX data.
result Margin requirements need to be significantly higher to reduce daily margin calls.
Study explores how neural networks and Transformers learn modular arithmetic with multiple inputs.
problem Understanding how neural networks and Transformers learn modular arithmetic with multiple inputs.
method Analytical characterization of features learned by neural networks and Transformers, focusing on margin maximization and Fourier spectra.
result Neural networks and Transformers require a minimum neuron count of \( m \geq 2^{2k-2} \cdot (p-1) \) to solve modular addition problems with \( k \) inputs and modulus \( p \).
Adversarial training is a principled approach for training robust neural networks. Despite of tremendous successes in practice, its theoretical properties still remain largely unexplored. In this paper, we provide new theoretical insights of gradient descent based adversarial training by studying its computational prop…
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.
Deep neural networks can generalize well even with perfect fits to noisy data.
problem Understanding the conditions under which deep neural networks generalize well in the presence of noise.
method Comprehensive study of linear maximum margin classifiers, focusing on noisy and noiseless cases.
result Discovery of a phase transition in test error bounds for the noisy model.
We define a generalized likelihood function based on uncertainty measures and show that maximizing such a likelihood function for different measures induces different types of classifiers. In the probabilistic framework, we obtain classifiers that optimize the cross-entropy function. In the possibilistic framework, we …
The paper explores how benign overfitting occurs in heavy-tailed input distributions.
problem Understanding overfitting in heavy-tailed input distributions.
method Analysis of maximum margin classifiers on unregularized logistic loss with gradient descent.
result Linear classifiers trained under certain conditions can asymptotically achieve the noise level as misclassification error.
Mirror flow optimizes separable data problems, converging to a maximum margin classifier.
problem Optimizing classification problems with separable data using mirror flow.
method Examine mirror flow on linearly separable classification problems, focusing on the horizon function of the mirror potential.
result Mirror flow converges to a maximum margin classifier for separable data under certain conditions.
Reduces quantifier variance with accuracy optimization of base classifier.
problem Minimizing quantifier variance under prior probability shift.
method Optimizes the Brier score of a base classifier for training data.
result Optimizing Brier score on training data reduces quantifier variance on test data.
The performance of a modulation classifier is highly sensitive to channel signal-to-noise ratio (SNR). In this paper, we focus on amplitude-phase modulations and propose a modulation classification framework based on centralized data fusion using multiple radios and the hybrid maximum likelihood (ML) approach. In order…
In a broad range of classification and decision making problems, one is given the advice or predictions of several classifiers, of unknown reliability, over multiple questions or queries. This scenario is different from the standard supervised setting, where each classifier accuracy can be assessed using available labe…
A fast method for training linear classifiers maximizes margins.
problem Training linear classifiers with maximum margins.
method Momentum-based gradient method derived from convex dual with Nesterov acceleration.
result Exponentially faster convergence rate compared to standard methods.
KS-algebra consists of expressions constructed with four kinds operations, the minimum, maximum, difference and additively homogeneous generalized means. Five families of Z-classifiers are investigated on binary classification tasks between English phonemes. It is shown that the classifiers are able to reflect well…
Multithreshold Entropy Linear Classifier (MELC) is a recent classifier idea which employs information theoretic concept in order to create a multithreshold maximum margin model. In this paper we analyze its consistency over multithreshold linear models and show that its objective function upper bounds the amount of mis…
A new method compares image classifiers using adaptive sampling of natural images.
problem Evaluation of image classifiers on small, fixed test sets may not generalize to real-world images.
method Adaptive sampling from a large corpus of unlabeled images to maximize classifier discrepancies measured by WordNet hierarchy.
result Human labeling of model-dependent image sets reveals relative classifier performance.
Campfire simplifies sparse training for CNNs, achieving high accuracy with minimal pruning.
problem Training CNNs with sparse weights to maintain high accuracy and robustness.
method Gradual pruning technique with simplified sparsity structure, exploring various tradeoffs.
result Sparse ResNet-50 and ResNet-50 v1.5 achieve 75% top-1 accuracy with 70% sparsity, <1% accuracy loss.
LC-CRFs are equivalent to HMMs, and MPM/MAP classifiers can be reformulated as CRFs.
problem Comparing and reformulating HMMs and CRFs.
method Demonstrating equivalence and reformulation of classifiers.
result LC-CRFs are equivalent to HMMs, and MPM/MAP classifiers can be reformulated as CRFs.
Adam optimizes linear classifiers with separable data.
problem Understanding Adam's implicit bias in linear logistic regression.
method Study of Adam's behavior on linearly separable data.
result Adam converges to a linear classifier with maximum ℓ∞-margin. Sparse multinomial logistic regression for multiclass classification with feature selection.
problem High-dimensional multiclass classification with a focus on sparse models.
method Penalized maximum likelihood with complexity penalty, feature selection using group Lasso and Slope classifiers.
result Achievement of minimax order in both small and large number of classes regimes.
We consider the orientation-preserving actions of finite groups G on pairs (S3,Γ), where Γ is a connected graph of genus g>1, embedded in S3. For each g we give the maximum order mg of such G acting on (S3,Γ) for all such Γ⊂S3. Indeed we will classify all graphs Γ⊂S3 which re…
New method for multiclass classification reduces error bounds.
problem High-dimensional multiclass classification with sparse coefficients.
method Sparse multinomial logistic regression with convex penalties.
result Plug-in classifiers achieve minimax generalization error bounds.
Label shift refers to the phenomenon where the prior class probability p(y) changes between the training and test distributions, while the conditional probability p(x|y) stays fixed. Label shift arises in settings like medical diagnosis, where a classifier trained to predict disease given symptoms must be adapted to sc…
Given a task of predicting Y from X, a loss function L, and a set of probability distributions Γ on (X,Y), what is the optimal decision rule minimizing the worst-case expected loss over Γ? In this paper, we address this question by introducing a generalization of the principle of maximum entropy. Applying t…
The paper classifies ruled surfaces in Lorentz-Minkowski space that are stationary for the moment of inertia.
problem Classifying ruled surfaces in Lorentz-Minkowski space that are stationary for the moment of inertia.
method Maximum principle applications, classification based on causal character of rulings.
result Planes are the only cylindrical stationary surfaces. For non-cylindrical surfaces, classification depends on the causal character of the rulings.
We consider orientation-preserving actions of finite groups G on pairs (S3,Σ), where Σ denotes a compact connected surface embedded in S3. In a previous paper, we considered the case of closed, necessarily orientable surfaces, determined for each genus g>1 the maximum order of such a G for all embeddings…
The paper analyzes the maximum margin algorithm's performance on noisy data.
problem Analyzing the performance of maximum margin algorithm on noisy data.
method Finite-sample analysis of maximum margin algorithm applied to noisy data.
result The maximum margin algorithm can achieve nearly optimal population risk with sufficient over-parameterization.
The paper compares one-hot encoding to Naïve Bayes for categorical variables.
problem Incorrect one-hot encoding affects Naïve Bayes performance.
method Mathematical and experimental analysis of PoB vs. categorical Naïve Bayes.
result Posterior probabilities are usually greater in the PoB case, but agree on the maximum a posteriori class label.
A new copula estimation method using classification.
problem Estimating copula density from joint and marginal distributions.
method Train a classifier to distinguish joint density from product of marginals.
result Empirically outperforms existing copula estimators.
Recently, a backdoor data poisoning attack was proposed, which adds mislabeled examples to the training set, with an embedded backdoor pattern, aiming to have the classifier learn to classify to a target class whenever the backdoor pattern is present in a test sample. Here, we address post-training detection of innocuo…
Study classifies translators for mean curvature flow in 3D.
problem Classifying semigraphical translators for mean curvature flow in R3. method Morse-Radó theory and angular maximum principle.
result No solution to the translator equation on the upper half-plane with alternating boundary values.
We consider the problem of classifying data manifolds where each manifold represents invariances that are parameterized by continuous degrees of freedom. Conventional data augmentation methods rely upon sampling large numbers of training examples from these manifolds; instead, we propose an iterative algorithm called M…
Paper introduces a novel method for estimating model confidence in deep neural classifiers.
problem Reliable confidence estimation for deep neural classifiers in safety-critical applications.
method Proposes a novel target criterion (true class probability) and learns it from data with an auxiliary model.
result The proposed method outperforms strong baselines in various tasks and network architectures.
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.
The paper classifies translation surfaces in a specific hyperelliptic component and finds the maximum number of disjoint geodesics.
problem Classifying and understanding translation surfaces in a specific hyperelliptic component.
method Analyzing geodesics and Euclidean structures on hyperelliptic surfaces.
result A classification theorem for translation surfaces in Hhyp(4) and the maximum number of disjoint geodesics. Associating distinct groups of objects (clusters) with contiguous regions of high probability density (high-density clusters), is central to many statistical and machine learning approaches to the classification of unlabelled data. We propose a novel hyperplane classifier for clustering and semi-supervised classificati…
Study improves adversarial classification using distributionally robust models.
problem Improving robustness against adversarial attacks in classification models.
method Distributionally robust chance constraints with Wasserstein ambiguity, reformulated as a regularized ramp loss minimization problem.
result Standard descent methods can converge to the global minimizer for the distributionally robust adversarial classification model.
Study axisymmetric surfaces in Euclidean space for energy minimization.
problem Finding surfaces in Euclidean space that minimize energy.
method Phase plane analysis and maximum principle.
result Helicoidal stationary surfaces must be rotational.
In this paper, we propose a general framework to learn a robust large-margin binary classifier when corrupt measurements, called anomalies, caused by sensor failure might be present in the training set. The goal is to minimize the generalization error of the classifier on non-corrupted measurements while controlling th…