Reformulation invariance on inference problems forces minimization of the Kullback-Leibler divergence, narrowing from f-divergences to alpha-divergences to KL.
hub
Iterating von Neumann’s procedure for extracting random bits
18 Pith papers cite this work. Polarity classification is still indexing.
18
Pith papers citing it
hub tools
citation-role summary
background 3
citation-polarity summary
verdicts
UNVERDICTED 18roles
background 3representative citing papers
Direct fixed-weight solver for free-support Wasserstein medians relocates atoms using OT barycentric projections and inverse-distance weights, achieving monotone descent on smoothed objectives with fewer subproblems than nested Weiszfeld baselines.
Introduces tangential Bayes denoiser for Riemannian Gaussian mixtures on manifolds via spectral Laplace-Beltrami approximation, with nearly Bayes risk in low noise and minimax optimality on the circle.
Characterizes the optimal e-power for ε-DP e-value hypothesis testing between P^n and Q^n, supplies a matching algorithm, and gives matching bounds on stopping times for private e-processes.
Primitive sequences obtained from iterated antiderivatives of the CDF are homeomorphic to probability measures on compact intervals, equivalent to factorial-rescaled moments of the reflected variable, and yield sharp bounds on functionals when the first m terms are fixed.
The profile maximum likelihood estimator for the location in anisotropic hyperbolic wrapped normal models is strongly consistent, asymptotically normal, and attains the Hájek-Le Cam minimax lower bound under squared geodesic loss.
Sufficient conditions are given for pseudo-likelihood estimation of both parameters in the Potts model at rate sqrt(N) for bounded-degree or irregular graphs, with impossibility shown for certain dense regular graphs, plus a new concentration inequality via nonlinear large deviations.
A new sampler for discrete distributions using coin flips that combines linearithmic space, negligible runtime overhead, and entropy-optimal expected flips in [H(P), H(P)+2).
Introduces the directional linear separability measure (LSM) as an asymmetric diagnostic for one-sided affine separability of neural representations.
A new test statistic and bootstrap for independence testing of high-dimensional nonstationary time series that avoids whitening by removing temporal dependence bias under the null.
Tri-SfSVD is a unified sparse functional SVD framework that performs simultaneous subject, feature, and temporal selection for biclustering and triclustering in longitudinal omics and EEG data.
Derives exact operating characteristic corrections and a numerical search over sample sizes to obtain optimal two-stage Bayes factor designs for two-arm binary-endpoint phase II trials that minimize expected sample size under the null.
Wasserstein least squares extends Euclidean least squares to distribution-valued responses via convex analysis, yielding n^{-1/2} rates under template deformation and faster barycenter rates than prior work.
LLMSurvival enables LLM-based survival analysis on tabular data by converting censored time-to-event tasks into pairwise comparisons, yielding small concordance gains over Cox and deep learning baselines on ICU mortality and fracture prediction.
Joint location-scale minimization for geometric medians on product manifolds degenerates to marginal medians, and three new scale-selection methods restore identifiability with asymptotic guarantees.
A review reframing density estimation as 'density evolution' across scales, linking kernel smoothing to heat flow, mixtures to compression, and topology to level sets, while stating three structural results on modes, Gaussian semigroups, and log-concavity.
A novel algorithm learns sets of optimal quantile regression trees to predict full conditional distributions interpretably and efficiently.
Proves reverse Poincaré inequality on global attractor of 2D reaction-diffusion system to obtain near-parametric statistical recovery of initial conditions from discrete observations.
citing papers explorer
-
Towards end-to-end LLM-based censoring-aware survival analysis
LLMSurvival enables LLM-based survival analysis on tabular data by converting censored time-to-event tasks into pairwise comparisons, yielding small concordance gains over Cox and deep learning baselines on ICU mortality and fracture prediction.