archive
Every paper Pith has read. Search by title, abstract, or pith.
883 papers in math.ST · page 9
-
Probabilistic algorithm recovers p-adic linear models from digit noise
$p$-adic Linear Regression for Random Sampling with Digitwise Noise
-
New algorithm fits linear models in p-adics under digit noise
$p$-adic Linear Regression for Random Sampling with Digitwise Noise
-
KL generalization error splits into model error
Information-Geometric Decomposition of Generalization Error in Unsupervised Learning
-
Seven-parameter model estimates dependent stress-strength reliability
Reliability estimation in dependent stress-strength model with Clayton copula and modified Weibull margins
-
QBIC selects true SEM for jump-diffusion processes
QBIC of SEM for jump-diffusion processes based on high-frequency data
-
NPMLE hits exact parametric rates when mixing distribution is finite
Adaptivity of the NPMLE to finitely discrete mixing distributions in Gaussian/Poisson mixtures
-
t-SNE converges to non-convex continuum limit with density penalty
On the continuum limit of t-SNE for data visualization
-
Sample weights locate change points in high-dimensional regression
Inferring Change Points in Regression via Sample Weighting
-
Traffic distributions determine GFOM dynamics on deterministic matrices
Universality of first-order methods on random and deterministic matrices
-
NetworkNet estimates nodal expansiveness and popularity
NetworkNet: A Deep Neural Network Approach for Random Networks with Sparse Nodal Attributes and Complex Nodal Heterogeneity
-
Polynomial branch-and-bound tree for least trimmed squares
Computation of Least Trimmed Squares: A Branch-and-Bound framework with Hyperplane Arrangement Enhancements
-
ADD achieves 100% accurate 48-bit watermarks resilient to distortions
ADD for Multi-Bit Image Watermarking
-
Single-parameter kernel method estimates average marginal effects in IV models
Average Marginal Effects in One-Step Partially Linear Instrumental Regressions
-
Random clock subordination adds jumps and memory to Wright-Fisher priors
Subordinated Wright-Fisher Priors
-
Nonparametric estimator tracks tail dependence in non-identical data
Trends in tail dependence of heteroscedastic extremes
-
Tensor factor models detect multiple change points and affected modes
Detection and Mode-Identification of Multiple Change Points in Tensor Factor Models
-
-
Pitman efficiency approximates relative efficiency under contamination
Empirical interpretation of the Pitman efficiency
-
α-MI derivative equals MMSE under tilted distributions on Gaussian channels
$\alpha$-Mutual Information for the Gaussian Noise Channel
-
Finite second moments are required for KL approximation by finite GMMs
Characterisations of Kullback--Leibler approximation by finite Gaussian mixtures
-
Diffusion samplers need Ω(√d) score queries
Query Lower Bounds for Diffusion Sampling
-
When the divergence function φ grows superlinearly
Sample Average Approximation for Distributionally Robust Optimization with $\phi$-divergences
-
Six-state Markov chain shows strict gap between compression methods
A Strict Gap Between Relaxed and Partition-Constrained Spectral Compression in a Six-State Lumpable Markov Chain
-
Trajectory regression hits optimal rate for SGD covariance
Online Covariance Estimation in Averaged SGD: Improved Batch-Mean Rates and Minimax Optimality via Trajectory Regression
-
Rank loss at coarser windows identifies exact depth in quiver Markov chains
Variable-Length Markov Chains on Finite Quivers: Boundary-Window Identifiability, Exact Depth, and Local Rank Comparison
-
Hidden entropy maximizer is the Markov extension of visible blocks
Entropy-Rate Selection for Partially Observed Processes
-
New bounds apply to heavy-tailed RLHF and SGLD
Tail-Aware Information-Theoretic Generalization for RLHF and SGLD
-
Half-normals disprove sum-max conjecture
A remark on the comparison of the sum and the maximum of positive random variables
-
Stationarity alone yields consistent Whittle estimation for Hawkes processes
Spectral analysis of multivariate stationary Hawkes processes
-
Debiasing produces valid confidence intervals for tensor estimates
Uncertainty Quantification for Noisy Low-tubal-rank Tensor Completion
-
Projected convergence on positive-measure directions implies weak convergence
Weak convergence from projected laws on a positive-measure set of directions
-
Multiway dependence yields normal limits for maximum score estimator at parametric rate
Gaussian approximation for maximum score and non-smooth M-estimators with multiway dependence
-
Normalized likelihood criteria recover true community count in sparse SBM
Normalized Likelihood Criteria for Model Selection in the Stochastic Block Model
-
Precision matrix gives Gaussian processes partial correlation networks
Partial correlation networks of Gaussian processes
-
Copula product unifies Wasserstein and rearranged dependence measures
On a copula product linking Wasserstein correlations and rearranged dependence measures
-
Beta process prior restores consistency for bivariate survival estimation
Bayesian bivariate survival estimation
-
Manifold regression recovers topology at optimal rate
Harmonic Map Regression: Rate-Optimal Nonparametric Estimation on Manifolds with Topological Recovery
-
Non-Gaussian densities estimated with far fewer samples
Data-Efficient Non-Gaussian Semi-Nonparametric Density Estimation for Nonlinear Dynamical Systems
-
Test for axial symmetry works with unknown direction
Testing axial symmetry around an unspecified direction
-
MoE models generalize like dense nets using only active parameters
Generalization and Scaling Laws for Mixture-of-Experts Transformers
-
Single known proxy identifies causal effects under completeness
Identifying Causal Effects Using a Single Proxy Variable
-
Tighter bounds on false discoveries for heterogeneous data
Confidence envelopes for the false discoveries with heterogeneous data
-
Logarithmic privacy budget is necessary for node-private SBM recovery
Node-Private Community Detection in Stochastic Block Models
-
Online method achieves optimal rate for quantile regression
Online Quantile Regression for Nonparametric Additive Models
-
Spectral test detects planted cliques from hypergraph matrix at √n
Planted clique detection and recovery from the hypergraph adjacency matrix
-
Fredriksson index vanishing marks benign overfitting
Spectral-Transport Stability and Benign Overfitting in Interpolating Learning
-
Entropy flow detects graph communities at 2% of Ricci flow's cost
An Efficient Entropy Flow on Weighted Graphs: Theory and Applications
-
BIC selects the right factor count even when models misspecify the data
Consistency of the Bayesian Information Criterion for Model Selection in Exploratory Factor Analysis
-
Adaptive 1-bit thresholds achieve order-optimal mean estimation
Order-Optimal Sequential 1-Bit Mean Estimation in General Tail Regimes
-
Adaptive 1-bit thresholds match unquantized mean rates for any k
Order-Optimal Sequential 1-Bit Mean Estimation in General Tail Regimes