Study tackles criterion collapse in learning criteria, showing conditions for loss minimization.
problem Criterion collapse in optimization, focusing on error probability minimizers.
method Analyzes various learning criteria, including DRO, OCE risks, and non-monotonic criteria.
result Non-monotonic criteria can avoid collapse, while monotonic ones cannot.
Hallucinations in models are mislinked estimates, not errors.
problem Hallucinations in generative models as failures to link estimates to plausible causes.
method Formalized hallucinations, showed even optimal estimators hallucinate, provided a general lower bound on hallucinate rate, reframed hallucination as structural misalignment, and experimentally supported theory.
result Hallucinations are structural misalignments between loss minimization and human-acceptable outputs, leading to estimation errors.
Consider the problem of minimizing functions that are Lipschitz and strongly convex, but not necessarily differentiable. We prove that after T steps of stochastic gradient descent, the error of the final iterate is O(log(T)/T) with high probability. We also construct a function from this class for which the error …
We address the problem of correcting group discriminations within a score function, while minimizing the individual error. Each group is described by a probability density function on the set of profiles. We first solve the problem analytically in the case of two populations, with a uniform bonus-malus on the zones whe…
TCE measures calibration error with a test-based approach.
problem Measuring calibration error of probabilistic binary classifiers.
method TCE uses a novel loss function based on a statistical test.
result TCE offers clear interpretation, consistent scale, and enhanced visual representation.
Probabilistic Autoencoder learns latent space weights' distribution.
problem Nonlinear model reconstruction error and sample quality.
method Normalizing flow for latent space weights' probability distribution.
result PAE achieves small reconstruction errors, high sample quality, and good performance.
The majority of traditional classification ru les minimizing the expected probability of error (0-1 loss) are inappropriate if the class probability distributions are ill-defined or impossible to estimate. We argue that in such cases class domains should be used instead of class distributions or densities to construct …
Improved analysis for diffusion models reduces KL divergence error dependence on data dimension and discretization step size.
problem Analyze the convergence of diffusion-based generative models under minimal assumptions.
method Model the generation process as a composition of reverse ODE and noising steps, leveraging Wasserstein-type error control and noise addition.
result Achieved a linear dependence on data dimension and improved dependence on discretization step size for KL divergence error.
New fairness concept extends minimax fairness to lexicographic fairness.
problem Fairness in supervised learning, especially lexicographic fairness.
method Introduced approximate lexifairness, derived algorithms for finding solutions, and proved generalization bounds.
result Proved that approximate lexifairness on training data implies approximate lexifairness on true distribution.
Study shows how neural networks generalize with minimal training data.
problem Understanding how neural networks generalize with limited data.
method Mean-field analysis of KL-regularized empirical risk minimization.
result Generalization error rate is O(1/n) for large n. Study shows benefits of transfer learning with neural networks.
problem Understanding generalization errors in transfer learning.
method Mean-field analysis applied to α-ERM and fine-tuning. result Established conditions for generalization error and convergence rates.
Establishes upper bounds on generalization error in active learning.
problem Improving query algorithms in active learning.
method Derives upper bounds on generalization error using informativeness and representativeness query strategies.
result Validates the use of regularization techniques to ensure bounds' validity.
In this paper, we present a novel sequential paradigm for classification in crowdsourcing systems. Considering that workers are unreliable and they perform the tests with errors, we study the construction of decision trees so as to minimize the probability of mis-classification. By exploiting the connection between pro…
The success of deep learning has led to a rising interest in the generalization property of the stochastic gradient descent (SGD) method, and stability is one popular approach to study it. Existing works based on stability have studied nonconvex loss functions, but only considered the generalization error of the SGD in…
We propose a robust risk measurement approach that minimizes the expectation of overestimation plus underestimation costs. We consider uncertainty by taking the supremum over a collection of probability measures, relating our approach to dual sets in the representation of coherent risk measures. We provide results that…
This work proves L2-regularized ERM controls smCE without post-hoc correction.
problem Calibration of predicted probabilities in machine learning models.
method Canonical L2-regularized empirical risk minimization. result Theoretical proof that smCE is controlled by ERM without post-hoc correction.
We generalize recent theoretical work on the minimal number of layers of narrow deep belief networks that can approximate any probability distribution on the states of their visible units arbitrarily well. We relax the setting of binary units (Sutskever and Hinton, 2008; Le Roux and Bengio, 2008, 2010; Montúfar and Ay,…
New policy minimizes error in finding best arm with privacy constraints.
problem Best arm identification with differential privacy constraints.
method Maximum absolute determinants principle for ε-DP policy. result Error probability decays exponentially in T with matching exponents. New truthful calibration errors improve model ranking in multiclass prediction.
problem Non-truthful calibration errors can mislead model comparisons.
method Introduced perfectly truthful calibration errors for multiclass predictions.
result Truthful calibration errors preserve decision-theoretic dominance and stabilize model rankings.
While the channel capacity reflects a theoretical upper bound on the achievable information transmission rate in the limit of infinitely many bits, it does not characterise the information transfer of a given encoding routine with finitely many bits. In this note, we characterise the quality of a code (i. e. a given en…
New algorithm controls type I error in NP classification under label noise.
problem Label noise affects NP classification methods, reducing power.
method Proposes a label-noise-adjusted Neyman-Pearson algorithm.
result Improves power while controlling type I error under desired level.
Adapts RKHS methods to estimate density ratios with optimal error.
problem Estimating density ratios from limited data.
method Minimizes regularized Bregman divergence in RKHS, with Lepskii type parameter choice.
result Adaptive minimax optimal error rate for quadratic loss.
Simpler majority vote of three classifiers achieves optimal error bounds.
problem Developing an optimal PAC learning algorithm in the realizable setting.
method Returning the majority vote of three ERM classifiers.
result Achieves optimal in-expectation bound on error.
PosCal training improves classification models by calibrating posterior probabilities.
problem Poorly calibrated posterior probabilities in classification models.
method End-to-end training procedure that directly optimizes the objective while minimizing the difference between predicted and empirical posterior probabilities.
result PosCal training achieves about 2.5% task performance gain and 16.1% calibration error reduction.
Estimating the largest community in a mixed population via sequential sampling.
problem Identifying the largest community in a mixed population with limited sampling.
method Sequential, random sampling of individuals across multiple boxes, optimizing sampling strategy and decision rule.
result Proposed algorithms achieve optimal error probability decay rates under fixed budget constraints.
The minimum error entropy (MEE) criterion has been verified as a powerful approach for non-Gaussian signal processing and robust machine learning. However, the implementation of MEE on robust classification is rather a vacancy in the literature. The original MEE only focuses on minimizing the Renyi's quadratic entropy …
Minimizes indecisions in selective classification to control misclassification rates.
problem Controlling misclassification rates in high-risk scenarios.
method Using indecisions to control misclassification rates, even below Bayes optimal.
result Control of misclassification rates to any user-specified level, even below Bayes optimal.
The paper analyzes greedy algorithms for MMD minimization, showing their efficiency and approximation error.
problem Minimizing Maximum Mean Discrepancy (MMD) for probability measure quantization.
method Iterative algorithms including kernel herding, greedy MMD minimization, and Sequential Bayesian Quadrature (SBQ).
result The greedy algorithms have a lower approximation error than SBQ, but are significantly faster.
This research improves deep neural network calibration using a new loss function.
problem Improving probability calibration in deep neural networks.
method Introduces Focal Calibration Loss (FCL) to minimize Euclidean norm and penalize calibration error.
result FCL achieves state-of-the-art performance in both calibration and accuracy metrics.
Decision makers increasingly rely on algorithmic risk scores to determine access to binary treatments including bail, loans, and medical interventions. In these settings, we reconcile two fairness criteria that were previously shown to be in conflict: calibration and error rate equality. In particular, we derive necess…
Robust hypothesis testing designs a test for worst-case distributions using kernel methods.
problem Design a robust test for hypothesis testing under uncertainty sets.
method Data-driven uncertainty sets constructed using kernel mean embeddings and maximum mean discrepancy (MMD). Bayesian and Neyman-Pearson settings investigated.
result Proposed robust kernel tests are exponentially consistent and asymptotically optimal.
New bounds for balanced classification improve understanding of imbalanced datasets.
problem Negligible size of the minority class in imbalanced datasets.
method Developed non-asymptotic and consistent bounds for balanced empirical risk minimization and balanced nearest neighbors estimates.
result Improved understanding of class-weighting benefits in real-world imbalanced classification settings.
In many signal detection and classification problems, we have knowledge of the distribution under each hypothesis, but not the prior probabilities. This paper is aimed at providing theory to quantify the performance of detection via estimating prior probabilities from either labeled or unlabeled training data. The erro…
Conditional restricted Boltzmann machines are undirected stochastic neural networks with a layer of input and output units connected bipartitely to a layer of hidden units. These networks define models of conditional probability distributions on the states of the output units given the states of the input units, parame…
Paper tackles MLR prediction error without assuming realizable models.
problem Prediction error in mixture of linear regressions without realizable assumptions.
method Developed algorithms for list-decoding MLR predictions and minimized empirical risk.
result Alternating minimization algorithm finds best fit lines in non-realizable settings.
For massive data, the family of subsampling algorithms is popular to downsize the data volume and reduce computational burden. Existing studies focus on approximating the ordinary least squares estimate in linear regression, where statistical leverage scores are often used to define subsampling probabilities. In this p…
Support vector machines (SVMs) naturally embody sparseness due to their use of hinge loss functions. However, SVMs can not directly estimate conditional class probabilities. In this paper we propose and study a family of coherence functions, which are convex and differentiable, as surrogates of the hinge function. The …
Study on error probability for classification of heavy-tailed renewal processes.
problem Error probability in classification of heavy-tailed renewal processes.
method Asymptotic expressions for Bhattacharyya bound on misclassification error probabilities.
result Obtained asymptotic expressions for misclassification error probabilities.
Study on error probabilities of machine learning classification techniques using large deviations theory.
problem Performance analysis of machine learning binary classification techniques.
method Large deviations theory applied to Data-Driven Decision Function (D3F) for error probability analysis.
result Classification error probabilities vanish exponentially, with an asymptotic formula providing precise error rate estimates.
The machine learning community has become increasingly concerned with the potential for bias and discrimination in predictive models. This has motivated a growing line of work on what it means for a classification procedure to be "fair." In this paper, we investigate the tension between minimizing error disparity acros…
A method for safe online classification reduces test costs while maintaining low error rates.
problem Sequential testing for binary disease outcomes with unknown logistic model parameters.
method Joint estimation of logistic parameter and feature distribution with a conservative threshold.
result Achieves target error with high probability and requires minimal excess tests.
We propose a novel neural sequence prediction method based on \textit{error-correcting output codes} that avoids exact softmax normalization and allows for a tradeoff between speed and performance. Instead of minimizing measures between the predicted probability distribution and true distribution, we use error-correcti…
A new one-step method for covariate shift adaptation.
problem Real-world data often violates the assumption of same distribution for training and test samples.
method Proposes a one-step optimization approach to jointly learn the model and weights.
result The proposed method achieves a generalization error bound and is empirically effective.
New framework minimizes interference and selection bias in network A/B testing.
problem Interference and selection bias in network A/B testing.
method Proposes a principled framework that jointly minimizes interference and selection bias using edge spillover probability and cluster matching.
result Significantly lower error in causal effect estimation compared to existing solutions.
The paper provides convergence bounds for approximating a distribution using point clouds.
problem Approximating a distribution using discrete points with minimal Wasserstein distance.
method Lloyd's algorithm with Power cells, analyzed using gradient descent.
result Explicit upper bounds for the convergence speed of the Lloyd-type algorithm.
Many machine learning tasks can be formulated as Regularized Empirical Risk Minimization (R-ERM), and solved by optimization algorithms such as gradient descent (GD), stochastic gradient descent (SGD), and stochastic variance reduction (SVRG). Conventional analysis on these optimization algorithms focuses on their conv…
Improved adaptive algorithms for identifying the best arm in MABs with fixed budget.
problem Identifying the best arm in stochastic Multi-Armed Bandits with a fixed sampling budget.
method Established a connection between Large Deviation Principles and adaptive algorithms, improving error probability bounds and devising new algorithms.
result The \sred algorithm outperforms existing algorithms in identifying the best arm.
Method curates cost-effective, high-quality datasets using AI models.
problem Costly manual labeling of datasets.
method Probably Approximately Correct Labels (PACL) method.
result Curates high-quality datasets with low overall labeling error.