Private and continual release of statistics
We ask the question: how can Web sites and data aggregators continually release updated
statistics, and meanwhile preserve each individual user's privacy? Suppose we are given a …
statistics, and meanwhile preserve each individual user's privacy? Suppose we are given a …
Minimax estimation of conditional moment models
We develop an approach for estimating models described via conditional moment
restrictions, with a prototypical application being non-parametric instrumental variable …
restrictions, with a prototypical application being non-parametric instrumental variable …
Efficient learning of generalized linear and single index models with isotonic regression
Abstract Generalized Linear Models (GLMs) and Single Index Models (SIMs) provide
powerful generalizations of linear regression, where the target variable is assumed to be a …
powerful generalizations of linear regression, where the target variable is assumed to be a …
Regularity as regularization: Smooth and strongly convex brenier potentials in optimal transport
FP Paty, A d'Aspremont… - … Conference on Artificial …, 2020 - proceedings.mlr.press
Estimating Wasserstein distances between two high-dimensional densities suffers from the
curse of dimensionality: one needs an exponential (wrt dimension) number of samples to …
curse of dimensionality: one needs an exponential (wrt dimension) number of samples to …
Silvar: Single index latent variable models
A semiparametric, nonlinear regression model in the presence of latent variables is
introduced. These latent variables can correspond to unmodeled phenomena or …
introduced. These latent variables can correspond to unmodeled phenomena or …
All your loss are belong to Bayes
Loss functions are a cornerstone of machine learning and the starting point of most
algorithms. Statistics and Bayesian decision theory have contributed, via properness, to elicit …
algorithms. Statistics and Bayesian decision theory have contributed, via properness, to elicit …
Error assessment of biogeochemical models by lower bound methods (NOMMA-1.0)
V Sauerland, U Löptien, C Leonhard… - Geoscientific Model …, 2018 - gmd.copernicus.org
Biogeochemical models, capturing the major feedbacks of the pelagic ecosystem of the
world ocean, are today often embedded into Earth system models which are increasingly …
world ocean, are today often embedded into Earth system models which are increasingly …
Learning to rank in supervised and unsupervised settings using convexity and monotonicity
S Acharyya - 2013 - repositories.lib.utexas.edu
This dissertation addresses the task of learning to rank, both in the supervised and
unsupervised settings, by exploiting the interplay of convex functions, monotonic mappings …
unsupervised settings, by exploiting the interplay of convex functions, monotonic mappings …
[PDF][PDF] Study on Approximate Gradient Projection (AGP) Property in Nonlinear Programming
R Sarin, BK Singh, S Rajan - International Journal of Mathematics …, 2014 - academia.edu
In this paper, we introduce an optimality condition that, roughly formulated, says that an
approximate gradient projection tends to zero. For this reason, we call it approximate …
approximate gradient projection tends to zero. For this reason, we call it approximate …
New designs to consistently estimate the isotonic regression
A Colubi, JS Dominguez-Menchero… - Computational …, 2018 - Springer
The usual estimators of the regression under isotonicity assumptions are too sensitive at the
tails. In order to avoid this problem, some new strategies for fixed designs are analyzed. The …
tails. In order to avoid this problem, some new strategies for fixed designs are analyzed. The …