#statistics

14 curated events

papersTODAY 04:00 UTC

Paper Explores When a General Factor Is Statistically Distinguishable

A new arXiv paper argues that whether an extra general dimension is needed beyond correlated first-order factors depends on the population covariance structure rather than on the estimator chosen. The authors show that a bifactor model is covariance-equivalent to a correlated-factors model under certain loading conditions, and they examine non-proportionality and structural stability as criteria. The work offers guidance for deciding when bifactor specifications are warranted.

psychometrics/factor-analysis/statistics/methodology

papersTODAY 04:00 UTC

Rater Ising-Potts Model Derives Weights From LLM Embeddings

A new arXiv paper introduces a Rater Ising-Potts model, an extension of the Ising model designed for multinomial rating data. The approach builds pairwise agreement indicators and category labels into the model, drawing its weights from large language model embeddings. The authors position the work as a link between network psychometrics and AI methods.

research/psychometrics/embeddings/statistics

papersTODAY 04:00 UTC

Functional SVD Framework Proposed for Regularized Multivariate Functional PCA

A new arXiv paper presents a framework for regularized multivariate functional principal component analysis built on a functional singular value decomposition. The approach generalizes existing MFPCA methods by adding dual penalization, which regularizes the decomposition in two ways. The work targets dimension reduction and analysis of multivariate functional data.

research/dimensionality-reduction/statistics/arxiv

papersTODAY 04:00 UTC

Density Ratio Estimation and Importance-Weighted Regression Under Target Shift

This paper examines how to estimate density ratios and perform importance-weighted regression when the output distribution shifts between training and test data while the conditional input distribution given outputs stays the same. The authors derive optimal estimation approaches for this continuous-output target shift setting. The work falls in the statistical machine learning area of distribution shift and covariate/label shift correction.

research/machine-learning/distribution-shift/statistics

papersSEP 11 04:00 UTC

arXiv paper proposes statistical guarantees for post-training hyperparameter selection

A new arXiv preprint addresses how to choose hyperparameters after a model has already been trained, such as inference-time settings and implementation-level options. The work aims to move this process beyond ad hoc tuning by providing statistical validity guarantees for the selected configuration. It targets deployment scenarios where pre-trained models still have multiple degrees of freedom that need to be fixed.

research/hyperparameters/post-training/statistics

papersSEP 10 04:00 UTC

Paper shows fixed-rollout pass@k evaluations identify only limited information

A new paper examines the common practice of extrapolating pass@k benchmark results to attempt counts larger than the number of samples actually collected per problem. Under a pooled conditional-Binomial model, the authors show that success counts from fixed-size rollouts determine only a finite number of distribution moments. The result implies such evaluations cannot fully characterize model performance well beyond the sampled regime.

evaluation/benchmarks/statistics/llm-evaluation

papersSEP 10 04:00 UTC

Statistical Method Proposed for Determining Sample Sizes in Machine Learning Prediction Models

Researchers have introduced a statistical framework for estimating how much data is needed to train machine learning prediction models. The approach addresses a key limitation of conventional power analysis, which normally requires the predictor-outcome relationship and effect structure to be defined in advance—something that is impractical for nonlinear models that learn complex patterns from data. The work appears in a new arXiv preprint filed under both artificial intelligence and machine learning categories.

machine-learning/statistics/methodology/research

papersSEP 10 04:00 UTC

Deep Fréchet Neural Network Framework Proposed for Metric-Space-Valued Regression

Researchers introduce DFNN, a deep neural network framework designed for regression tasks where the response variable lies in a general metric space rather than Euclidean coordinates. The approach targets non-Euclidean outputs such as probability distributions, networks, symmetric positive-definite matrices, and compositional data. It extends Fréchet regression concepts into deep learning architectures for these increasingly common data types.

deep-learning/regression/statistics/metric-spaces

papersSEP 10 04:00 UTC

Normalizing flow decomposition enables likelihood-free inference with nuisance parameters

A new arXiv paper proposes splitting a neural normalizing flow into components that expose a quantity close to a pivotal statistic when nuisance parameters are present. The approach requires only a sample generator from the target distribution rather than an explicit likelihood. This could simplify inference in settings where the likelihood is intractable but simulation is straightforward.

normalizing-flows/statistics/inference/research

papersSEP 10 04:00 UTC

Balanced k-shot sampling causes exact degeneracy in discriminant analysis on LLM embeddings

A new arXiv paper proves that balanced k-shot sampling, which draws exactly k labeled examples per class, induces an exact and provable degeneracy in a family of small-sample discriminant estimators. The result concerns kernelized linear discriminant methods applied to LLM embeddings, where the within-class scatter operator breaks down under equal per-class sample counts. The finding carries practical consequences for few-shot classification pipelines that rely on embeddings from large language models.

llm-embeddings/few-shot/statistics/classification

papersSEP 10 04:00 UTC

arXiv paper offers unifying perspective on probabilities as model predictions

A new cs.LG preprint tackles the long-standing philosophical divide between Bayesian and frequentist interpretations of probability. The authors propose framing probabilities as predictions generated by models, aiming to reconcile competing viewpoints. The work also examines under what conditions acting on probabilistic claims produces desirable outcomes.

machine-learning/statistics/theory/bayesian

papersSEP 10 04:00 UTC

Study Establishes Gaussian Approximation Bounds for Martingale Sums from Ergodic Markov Chains

A new paper posted to arXiv derives Gaussian approximation bounds, measured in higher-order Wasserstein distance, for sums of multivariate martingale differences produced by uniformly ergodic Markov chains. The results rely on an L^(2+η)p moment condition with η>0, a modest strengthening of standard integrability requirements. The work was announced as a cross-listing to the machine learning category, reflecting its potential relevance to statistical analysis of dependent data.

statistics/probability-theory/markov-chains/martingales

papersSEP 10 04:00 UTC

Regularized Estimation and Feature Selection in Mixtures of Generalized Linear Experts

A revised arXiv paper studies mixtures of experts, conditional mixture models in which both the mixing weights and component densities depend on predictors. The work develops regularized estimation methods with feature selection for mixtures of generalized linear experts, supporting regression, classification, and model-based clustering of heterogeneous data.

mixture-of-experts/feature-selection/statistics/regression

papersSEP 10 04:00 UTC

MiNCE: Consistent Confidence Envelopes for Band-Limited Functions Introduced in New Paper

A new arXiv preprint presents MiNCE, a minimum-norm method for constructing nonparametric, simultaneous confidence regions around band-limited functions and their smoothed spectra. The approach draws on reproducing kernel Hilbert space theory to deliver confidence envelopes that are valid for finite samples and provably strongly consistent.

research/uncertainty/statistics/rkhs