archive
Every paper Pith has read. Search by title, abstract, or pith.
6894 papers in stat · page 4
-
Minimax formulas derived for spectral densities in random-field estimation
Minimax approach to the estimation problem for homogeneous random fields
-
Sequential shrinkage attains oracle risk for multi-source data
Tuning-Free Efficient Estimation for Multi-Source Data via Covariance-Aware Shrinkage
-
Any valid transport map is as hard to estimate as the OT map
The Fundamental Limits of Valid Transport Map Estimation
-
Continual learning reduces to sequential projections in homogeneous nets
Convergence of Continual Learning in Homogeneous Deep Networks
-
Faster updates align covariance matrices under Wasserstein distance
ITSPACE: Monotone Gaussian Optimal Transport Updates
-
DR-ACI gives prediction intervals for causal effects under time dependence
Doubly Robust Adaptive Conformal Inference for Causal Effects Under Temporal Dependence
-
Factorizable flows learn each parameter's density effect in isolation
Factorizable Normalizing Flows for parameter-dependent density morphing
-
Causal drift recovered from equilibrium snapshots
Non-parametric recovery of causal diffusion mechanisms from steady-state observations
-
Curvature-weighted noise halves SGD error floor on quadratics
Curvature-Weighted Gradient Diversity: A Noise Measure for Geometry-Adaptive SGD Schedules
-
SGD learns spurious shortcut exponentially fast before XOR signal
SGD Provably Prioritizes a Shortcut Spurious Feature in the XOR Model
-
MLE of coupling parameter consistent and efficient in hidden OU process
Parameter estimation in a fully coupled partially observed Ornstein-Uhlenbeck process
-
Rademacher max-average expectation bounded below by min{255/256
Notes on constants for maxima of Rademacher averages
-
Net-LSW tracks time-varying dependencies on networks
Multiscale Dynamic Dependence Estimation over Networks
-
Shell-core geometry produces grokking scaling laws
A Stochastic--Geometric Theory of Scaling Laws in Grokking
-
Horseshoe prior controls FDR at optimal detection boundary
Multiple testing with the horseshoe
-
Horseshoe posterior rules attain optimal detection boundary
Multiple testing with the horseshoe
-
Extrapolation beats single-nugget solves for ill-conditioned systems
Extrapolating from Regularised Solutions for Solving Ill-Conditioned Linear Systems in Machine Learning
-
Conditional test unifies HWE check and SNP association in GWAS
Evaluating HWE and Association in Genome Wide Association Studies: A Unified Procedure
-
CDFs replace sorting for sliced Wasserstein distance
Highly Data Parallelizable Estimation of the Sliced-Wasserstein Distance Using Cumulative Distribution Functions
-
Differential algebra checks unique recovery of functions in DE models
Structural functional identifiability and model discovery in differential equation models
-
Level sets yield exact acceptance certificates for speculative decoding
When Is a Draft Accepted? A Theory of Acceptance in Speculative Decoding
-
TabPFN v2 leads NHANES benchmark for HbA1c and CRP
Accelerometry-Derived Digital Biomarkers for Cardiometabolic Risk: A Population-Representative Tabular Benchmark with Uncertainty Quantification
-
Optimal confidence sets induce Choquet-risk optimal contours
Efficiency of Valid Inferential Models: Choquet-risk Optimal Possibility Measures, and Direct Comparisons
-
Optimal transport unifies flow matching and Schrödinger bridge
Notes on generative modeling: flow matching, diffusion, optimal transport and Schr{\"o}dinger bridge
-
Copulas obey |β|^3 ≤ 2ξ exactly for Chatterjee ξ and Blomqvist β
The exact region between Chatterjee's $\xi$ and Blomqvist's $\beta$
-
Explicit MSE bounds for adaptive rare MCMC under Wasserstein contraction
Error bounds for simultaneous Wasserstein contractive adaptive increasingly rare MCMC
-
Common noise repeats speed nonparametric regression rates
Adaptive nonparametric regression from repeated measurements under common noise
-
Frank-Wolfe computes optimal e-values for non-convex voting tests
Optimal Posterior E-values with Non-Convex Parameter Sets with Applications to Voting Systems
-
Ordinal time series model estimates category spacings from data
Beyond Equidistant Assumptions: An Autoregressive Ordered Stereotype Model for Ordinal Time Series
-
Four Lorenz curve forms fail validity conditions
Revisiting "A universal model for the Lorenz curve with novel applications''
-
Personal decision theory optimal under population enforcement metric
A causal modeling perspective on decision theory
-
AdaGrad misses Hölder rate on composite problems
AdaGrad does not adapt to H\"older-smoothness for composite objectives
-
Shapley values track decision value not forecast error
Decision-Value Attribution in Predict-then-Optimize Systems
-
Squeals flag poor model fits while users drag curves
The Squealer: Sensification of model exploration and model misfit
-
Exact privacy math for 2020 Census now 1824 times faster
A Sieve-Accelerated Quadrature Method for Exact Privacy Accounting in the 2020 U.S. Decennial Census
-
New downscaling method matches kriging accuracy at far lower compute cost
Scalable coarse-to-fine spatial downscaling
-
Autoencoder memorizes inliers before outliers in early training
What Drives the Inlier-Memorization Effect? A Theory of Outlier Detection via Early Training Dynamics
-
Historical data cuts bias and variance in model evaluations
HERO: Improving the Reliability and Sensitivity of Generative Model Evaluation Using Historical Data
-
Statistics students should probe LLMs as stochastic systems
Probing the Stochastic Machine: Engaging with LLMs in Statistics Curricula Through Veridical Data Science
-
Symmetric noise enables valid tests after data-driven selection
Testing hypotheses via orthogonalization
-
Distance matrix yields latent manifold dimension from eigenvalue multiplet
I-BBS: Coordinate-Free Inference of Latent Sub-Manifolds Using Random Distance Matrix Theory
-
Adjusted Wasserstein distance improves MDS on heavy-tailed data
Adjusted Wasserstein distances for bridging empirical and true distributions with applications to MDS
-
Summary statistics enable privacy-preserving transfer for single-index models
Multi-Source Transfer Learning of Sparse Single-Index Models
-
Shared blocks enable consistent class graph recovery
Beyond Local Independence: High-Dimensional Latent Class Graphical Models with Shared Block Structure
-
Malliavin weights cut fade-probability variance by up to 2516x
Stochastic Analysis of Fade Duration Using Wiener Chaos Expansion and Malliavin Calculus: Optimal Importance Sampling via Adaptive SGD
-
Bidirectional diffusion checks MHD prediction errors without ground truth
Bidirectional Autoregressive Latent Diffusion for Forward and Inverse Magnetohydrodynamics
-
AI settles worst-case complexity of 1937 Kaczmarz algorithm
How AI settled the complexity of the oldest SGD algorithm
-
Optimizer memory makes shuffle order first-order fine-tuning noise
Optimizer Memory Makes Shuffle Order a First-Order Source of Fine-Tuning Noise
-
Mixture models distinguish mild from gross anomalies in circular data
Modelling and detecting mild and gross anomalies in circular data via double-contaminated models
-
Gradient tweak keeps primary descent intact while meeting secondary goals
Not All Objectives Are Born Equal: Priority-Constrained Descent for Hierarchical Multi-Objective Optimization