On the tight constant in the multivariate Dvoretzky–Kiefer–Wolfowitz inequality

@article{Naaman2021OnTT,
  title={On the tight constant in the multivariate Dvoretzky–Kiefer–Wolfowitz inequality},
  author={Michael Naaman},
  journal={Statistics \& Probability Letters},
  year={2021},
  volume={173},
  pages={109088}
}
  • M. Naaman
  • Published 6 March 2021
  • Mathematics
  • Statistics & Probability Letters

FALSE DISCOVERY RATE CONTROL WITH UNKNOWN NULL DISTRIBUTION: IS IT POSSIBLE TO MIMIC THE ORACLE? BY ETIENNE ROQUAIN

Classical multiple testing theory prescribes the null distribution, which is often too stringent an assumption for nowadays large scale experiments. This paper presents theoretical foundations to

Non-asymptotic confidence bands on the probability an individual benefits from treatment (PIBT)

The premise of this work, in a vein similar to predictive inference with quantile regression, is that observations may lie far away from their conditional expectation. In the context of causal

False discovery rate control with unknown null distribution: Is it possible to mimic the oracle?

Classical multiple testing theory prescribes the null distribution, which is often a too stringent assumption for nowadays large scale experiments. This paper presents theoretical foundations to

Random Search Hyper-Parameter Tuning: Expected Improvement Estimation and the Corresponding Lower Bound

TLDR
An empirical estimate for the expected accuracy improvement from an additional iteration of hyperparameter search is established and is bound with an error of O (cid:18)q log kk ( cid:19) w.p.h. where k is the current number of iterations.

Bilateral Trade: A Regret Minimization Perspective

TLDR
This paper casts the bilateral trade problem in a regret minimization framework over ) rounds of seller/buyer interactions, with no prior knowledge on their private valuations, and proves the following tight bounds on the regret.

A Reduction to Binary Approach for Debiasing Multiclass Datasets

TLDR
It is proved that R2B satisfies optimality and bias guarantees and empirically that it can lead to an improvement over two baselines: treating multiclass problems as multi-label by debiasing labels independently and transforming the features instead of the labels.

A General Framework for Powerful Confounder Adjustment in Omics Association Studies

Genomic data are subject to various sources of confounding, such as batch effects and cell mixtures. To identify genomic features associated with a variable of interest in the presence of confounders,

ECOD: Unsupervised Outlier Detection Using Empirical Cumulative Distribution Functions

TLDR
A novel outlier detection method called ECOD (Empirical-Cumulative-distribution-based Outlier Detection), which is inspired by the fact that outliers are often the “rare events” that appear in the tails of a distribution.

A universal probabilistic spike count model reveals ongoing modulation of neural variability

TLDR
A universal probabilistic spike count model is presented that defies a simple parametric relationship with mean spike count as assumed in standard models, its modulation by external covariates can be comparably strong to that of the mean firing rate, and slow low-dimensional latent factors explain away neural correlations.

MLDemon: Deployment Monitoring for Machine Learning Systems

TLDR
A novel approach, MLDemon, which integrates both unlabeled data and a small amount of on-demand labels to produce a real-time estimate of the ML model’s current performance on a given data stream and is theoretical analysis to show that MLDemon is minimax rate optimal for a broad class of distribution drifts.

References

SHOWING 1-10 OF 16 REFERENCES

The Tight Constant in the Dvoretzky-Kiefer-Wolfowitz Inequality

Let F^ n denote the empirical distribution function for a sample of n i.i.d. random variables with distribution function F. In 1956 Dvoretzky, Kiefer and Wolfowitz proved that P(√n sup x (F^ n

Bernstein inequality and moderate deviations under strong mixing conditions

In this paper we obtain a Bernstein type inequality for a class of weakly dependent random variables. The proofs lead to a moderate deviation principle for sums of bounded random variables with

Inequalities and limit theorems for weakly dependent sequences

These notes are a translation into Englsh of the preprint "Theoremes limites pour les suites de variables aleatoires faiblement dependantes", Prepublication 97-81 de l'Universite de Paris-Sud" which

Weak Convergence of Probabilities on Nonseparable Metric Spaces and Empirical Measures on Euclidean Spaces

It is known that under certain mild set-theoretic assumptions, a finite, countably additive measure defined on all Borel sets of a metric space is concentrated in a separable subspace (Marczewski and

The Dantzig selector: Statistical estimation when P is much larger than n

TLDR
The main point of the paper is accurate statistical estimation in high dimensions, which includes theoretical, practical and computational issues, and the Dantzig Selector compares with the Lasso.

Tail Behaviour for Suprema of Empirical Processes

Abstract : This document considers multi-variate empirical processes with an empirical distribution function based on i.i.d. variables with certain distribution functions.

Approximation Theorems of Mathematical Statistics

Preliminary Tools and Foundations. The Basic Sample Statistics. Transformations of Given Statistics. Asymptotic Theory in Parametric Inference. U--Statistics. Von Mises Differentiable Statistical